Upload
others
View
11
Download
1
Embed Size (px)
Citation preview
2017
TRC1501
Performance of Asphalts Modified with
Polyphosphoric Acid
Zahid Hossain, Andrew F. Braham, Gaylon Baumgardner
Final Report
TRC 1501 Final Report
Page | 1
Performance of Asphalts Modified with Polyphosphoric Acid
Final Report
TRC 1501
By
Zahid Hossain, Ph.D., P.E.
Associate Professor
Department of Civil Engineering
Arkansas State University
Andrew F. Braham, Ph.D., P.E.
Associate Professor
Department of Civil Engineering
University of Arkansas
Gaylon Baumgardner, Ph.D.
Executive Vice President
Paragon Technical Services, Inc.
December 2017
The statements contained in this report are the sole responsibility of the authors, and do not
necessarily reflect the official views of the Arkansas State Highway and Transportation
Department. The use of specific product names in this paper does not constitute or imply an
endorsement of those products. This document does not constitute a standard, specification, or
regulation.
TRC 1501 Final Report
Page | 2
ABSTRACT
Among the three polymer additives approved by Arkansas Department of Transportation
(ArDOT), styrene butadiene styrene (SBS) is the widely used co-polymer. Like many other
states, Polyphosphoric Acid (PPA) is not an approved modifier in Arkansas. Even though
ArDOT does not allow the use of PPA in asphalt, some suppliers may use it due to its
economical advantages. The objective of this study was to find out short-term and long-term
performance of asphalts modified with PPA. To evaluate the performance of PPA-modified
binders, mechanistic, chemical and moisture susceptibility tests were conducted in laboratories
on both asphalt binders and asphalt mixtures. To this end, base Performance Grade (PG) binders
from two different crude sources were modified with different dosages of PPA along with its
polymer counterpart, SBS. The mechanistic performance was evaluated by conducting
Superpave tests, chemical properties were determined through acid detection and pH
measurements, Fourier Transform Infrared Spectroscopy (FTIR) analysis, and saturates,
aromatics, resins, and asphaltenes (SARA) analysis. Asphalt mixtures prepared from PPA- and
SBS-modified binders and a selected aggregate were tested for moisture resistance, rutting, and
creep behavior. Further, binders recovered from core samples obtained from old roadway
sections (good and poor performing) were tested to ascertain if PPA had any adverse effects on
them. The moisture susceptibility was predicted by employing the Surface Free Energy (SFE)
technique. Superpave test results of asphalt binders suggest that PPA-modified binders are more
rutting, fatigue cracking, and low temperature cracking resistant than PG 64-22 binders, but the
use of Liquid Anti-Stripping Agents (LAA) may deteriorate the rutting performance. The SFE
analysis did not show any significant negative effects on stripping potential of PPA-modified
binder and aggregate systems. Asphalt mixture performance test data was found to be in
agreement with the binder test data. The acid detection test method (AASHTO TP 78) is
recommended to be a quick and easy test method for detecting the presence of PPA. Overall,
there were no significant downsides to using PPA, but the PPA did not behave identically across
asphalt binders from different sources. Findings of this study are expected to be helpful in
revising the ArDOT specifications regarding PPA modification of asphalts.
TRC 1501 Final Report
Page | 3
Table of Contents
1. Introduction and Problem Statement .................................................................................... 10
2. Background and Literature Review ...................................................................................... 11
3. Research Objectives .............................................................................................................. 20
4. Research Methodologies ....................................................................................................... 21
4.1 Survey Neighboring States ............................................................................................. 21
4.2 Survey Certified Asphalt Binder Suppliers in Arkansas ............................................... 26
4.3 Test Plan ......................................................................................................................... 29
4.4 Test Methods .................................................................................................................. 33
4.4.1 Asphalt Binder Rheological Tests .......................................................................... 33
4.4.2 Chemical Performance Tests .................................................................................. 40
4.4.3 Asphalt Mixture Tests ............................................................................................. 48
4.5 Field Performance Data Collection ................................................................................ 54
5 Data and Analysis ................................................................................................................. 57
5.1 Binder Performance Tests .............................................................................................. 57
5.1.1 Penetration Test Results .......................................................................................... 57
5.1.2 Rotational Viscosity (RV) Test Results .................................................................. 58
5.1.3 Dynamic Shear Rheometer (DSR) Test Results ..................................................... 60
5.1.4 Bending Beam Rheometer (BBR) Test Results ...................................................... 67
5.1.5 Moisture Susceptibility Analysis Using Surface Free Energy (SFE) Analysis ...... 70
5.1.6 Elastic Recovery (ER) and Multiple Stress Creep Recovery (MSCR)……………78
5.2 Chemical Test Results .................................................................................................... 79
5.2.1 Acidity Measurement Test ...................................................................................... 79
5.2.2 Saturates Aromatics Resins and Asphaltenes (SARA) Analysis ............................ 80
5.2.3 Fourier Transform Infrared Spectroscopy (FTIR) .................................................. 82
5.3 Mixture Performance Tests ............................................................................................ 87
5.3.1 Evaluation of Rutting and Stripping of Asphalt (ERSA) Test ................................ 87
5.3.2 Uniaxial Dynamic Modulus .................................................................................... 88
5.3.3. Semi-Circular Bend (SCB) Fracture Test ............................................................... 90
5.3.4 IDT Creep Compliance and Indirect Tensile Strength ........................................... 91
5.3.5 Tensile Strength Ratio............................................................................................. 92
5.4 Field Performance Data Collection ................................................................................ 93
TRC 1501 Final Report
Page | 4
5.4.1 Acid Detection Test ................................................................................................ 93
5.4.2 MMHIS Results .......................................................................................................... 94
5.4.3 Summary of Field Performance Data ......................................................................... 95
6. Conclusions and Recommendations ......................................................................................... 96
6.1 Conclusions .................................................................................................................... 96
6.2 Recommendations .......................................................................................................... 97
6.3 Implementaiton............................................................................................................. 978
7 Acknowledgements ............................................................................................................... 99
8 References ........................................................................................................................... 100
TRC 1501 Final Report
Page | 5
List of Tables
Table 4.1: Certified Asphalt Binder Suppliers in Arkansas .......................................................... 27
Table 4.2: Details of Sample Nomenclature ................................................................................. 31
Table 4.3: Nomenclatures of LAA Modified Binders .................................................................. 32
Table 4.4: Superpave Specification for Rutting and Fatigue Factor ............................................. 36
Table 4.5: Superpave Specification for BBR Test ........................................................................ 37
Table 4.6: Specific Gravity of Coarse Aggregate ......................................................................... 52
Table 4.7: Specific Gravity of Fine Aggregate ............................................................................. 53
Table 4.8: Bulk Specific Gravity (Gmb) at Ndesign of 75 .............................................................. 53
Table 4.9: Bulk Specific Gravity (Gmb) at Nmax of 115 ............................................................... 54
Table 4.10: Bulk Specific Gravity (Gmm) ................................................................................... 54
Table 4.11: Threshold Values of Pavement Distresses for ArDOT.............................................. 55
Table 5.1: Rotational Viscosity (mPa.s) Data of S1 and S2 samples ........................................... 59
Table 5.2: Mixing and Compaction Temperatures of PPA and SBS Modified Binders .............. 60
Table 5.3: SFE Parameters (mJ/m2) and Cohesion Energy (mJ/m
2) of asphalt binders ............... 72
Table 5.4: Work of Adhesion (mJ/m2) for Asphalt-Aggregate System in Dry Condition ........... 73
Table 5.5: Work of Adhesion (mJ/m2) for Asphalt-Aggregate System in Wet Condition ........... 74
Table 5.6: Qualitative Description of Compatibility .................................................................... 75
Table 5.7: Compatibility Ratio of PPA+LAA Modified Binders ................................................. 78
Table 5.8: Absorbance and Area Analysis of S2B7 and S2B8 ..................................................... 86
Table 5.9: Evaluation of Rutting and Stripping of Asphalt Mixtures ........................................... 88
Table 5.10: Acid Detection Test Sample Details for Field Performance Data Collection ........... 94
TRC 1501 Final Report
Page | 6
List of Figures
Figure 4.1: Respondents of State Survey, ..................................................................................... 22
Figure 4.2: States Allowing PPA as a Modifier in Asphalt Binder. ............................................. 22
Figure 4.3: Existing Specifications Regarding PPA Modification. .............................................. 23
Figure 4.4: Concerns on Using PPA in Asphalt Binders. ............................................................. 24
Figure 4.5: Using PPA-modified binders in Last Five Years. ...................................................... 24
Figure 4.6: Response Whether Contractors Notify when PPA-modified Asphalt Binders is Used.
....................................................................................................................................................... 25
Figure 4.7: Popularity of Different Polymers Other Than PPA for Modifying Asphalt Binders. 26
Figure 4.8: Summary of Research Methodology. ......................................................................... 30
Figure 4.9: Penetration Test Device. ............................................................................................. 33
Figure 4.10: RV Test Device. ....................................................................................................... 34
Figure 4.11: Dynamic Shear Rheometer. ...................................................................................... 35
Figure 4.12: Bending Beam Rheometer (BBR). ........................................................................... 37
Figure 4.13: Rotational Thin Film Oven (RTFO). ........................................................................ 38
Figure 4.14: Pressure Aging Vessel (PAV). ................................................................................. 39
Figure 4.15: Vacuum Degassing Oven. ........................................................................................ 39
Figure 4.16: Stretching and Bending Vibrations of Atoms due to Absorption of Infrared
Radiation. ...................................................................................................................................... 41
Figure 4.17: Experimental Set-up for FTIR Spectroscopy. .......................................................... 41
Figure 4.18: An Empty IR card (Left) and a Sample Ready for FTIR Test (Right). .................... 43
Figure 4.19: Acid Detection Test Results. .................................................................................... 44
Figure 4.20: Optical Contact Analyzer (OCA) Device. ................................................................ 45
Figure 4.21: AN asphalt Binder Sample Ready for OCA Test. .................................................... 47
TRC 1501 Final Report
Page | 7
Figure 4.22: A Typical Rock Sample After Cutting. .................................................................... 47
Figure 4.23: Left and Right ERSA Samples to Perform Test. ...................................................... 49
Figure 4.24: Dynamic Modulus in IDT Configuration. ................................................................ 50
Figure 4.25: Uniaxial Dynamic Modulus. .................................................................................... 51
Figure 4.26: SCB Test................................................................................................................... 51
Figure 4.27: Automated Road Analyzer ....................................................................................... 55
Figure 4.28: MMHIS Screenshot. ................................................................................................. 56
Figure 5.1: Penetration Test Results. ............................................................................................ 58
Figure 5.2: Penetration Test Results for PPA+LAA Modified Samples. ..................................... 58
Figure 5.3: DSR Test Results of Unaged Binders from Source 1. ............................................... 61
Figure 5.4: DSR Test Results of Unaged Binders from Source 2. ............................................... 61
Figure 5.5: DSR Test Results of RTFO-aged Binders from Source 1. ......................................... 62
Figure 5.6: DSR Test Results of RTFO-aged Binders from Source 2. ......................................... 62
Figure 5.7: DSR Test Results of Unaged PPA+LAA Modified Binders from Source 1. ............. 63
Figure 5.8: DSR Test Results for Unaged PPA+LAA Binders from Source 2. ........................... 64
Figure 5.9: DSR Test Results of RTFO-Aged PPA+LAA Binders from Source 1. ..................... 64
Figure 5.10: DSR Test Results of RTFO-Aged PPA+LAA Binders from Source 2. ................... 65
Figure 5.11: DSR Test Results of PAV-aged Binders from Source 1. ......................................... 66
Figure 5.12: DSR Test Results of PAV-aged Binders from Source 2. ......................................... 66
Figure 5.13: DSR Test Results of PAV-aged PPA+LAA Binders from S1. ................................ 67
Figure 5.14: DSR Test Results of PAV-aged PPA+LAA Binders from S2. ................................ 67
Figure 5.15: Creep Stiffness of Binders from Source 1. ............................................................... 68
Figure 5.16: Creep Stiffness of Binders from Source 2. ............................................................... 68
Figure 5.17: “m-values” of Binders from Source 1. ..................................................................... 69
TRC 1501 Final Report
Page | 8
Figure 5.18: “m-values” of Binders from Source 2. ..................................................................... 69
Figure 5.19: Contact Angles of Binders from Source 1................................................................ 71
Figure 5.20: Contact Angles of Binders from Source 2................................................................ 71
Figure 5.21: Compatibility Ratio of Binders from Source 1. ....................................................... 76
Figure 5.22: Compatibility Ratio of Binders from Source 2. ....................................................... 76
Figure 5.23: Variation of CR Values with Different Dosage of PPA for Source 1. ..................... 77
Figure 5.24: Variation of CR Value with Different Dosage of PPA for Source 2. ...................... 77
Figure 5.25: pH of Asphalt Binders. ............................................................................................. 80
Figure 5.26: SARA Fractions of Source 1 Asphalt Binders. ........................................................ 81
Figure 5.27: SARA Fractions of Source 2 Asphalt Binders. ........................................................ 81
Figure 5.28: FTIR Spectrum of PG 64-22 Binders from S1 and S2. ............................................ 82
Figure 5.29: FTIR Spectra of PPA-modified Binders from Source 1. .......................................... 83
Figure 5.30: FTIR Spectra of PPA-modified Binders from Source 2. .......................................... 84
Figure 5.31: FTIR Spectrum for PPA+LAA binders from Source 1. ........................................... 84
Figure 5.32: FTIR Spectrum for PPA+LAA Modified Binders from Source 2. .......................... 85
Figure 5.33: Polymer Content Analysis of S2B7. ........................................................................ 85
Figure 5.34: Polymer Content Analysis of S2B8. ........................................................................ 86
Figure 5.35: Summary of ERSA Test Results. ............................................................................. 88
Figure 5.36: Uniaxial Dynamic Modulus Master Curves Summary. ........................................... 89
Figure 5.37: Shift Factor – Uniaxial Dynamic Modulus. ............................................................. 90
Figure 5.38: Summary of SCB Test Fracture Energy at -10°C..................................................... 91
Figure 5.39: Summary of SCB Test Fracture Energy at -24°C..................................................... 91
Figure 5.40: Summary of IDT Tensile Strength at -10°C. ............................................................ 92
Figure 5.41: Summary of Tensile Strength Ratio. ........................................................................ 93
TRC 1501 Final Report
Page | 9
Figure 5.42: Average IRI Values for Mixture Containing PPA-modified Binder. ....................... 95
Figure 5.43: Average Rutting Data for Mixture Containing PPA. ............................................... 95
TRC 1501 Final Report
Page | 10
1. Introduction and Problem Statement
According to the 2014 Standard Specifications for Highway Construction of the Arkansas State
Highway and Transportation Department (ArDOT), three types of Performance Grade (PG)
binders (PG 64-22, PG 70-22, and PG 76-22) are approved to be used in highway construction
projects in Arkansas. It further states that PG 70-22 and PG 76-22 binders shall be production
straight run binders that are modified by using either a SB, SBS or SBR to achieve the specified
grade. Although the ArDOT does not approve PPA as a modifier of asphalt binder, some
suppliers may use it to get the desired Performance Grade (PG). This is mainly due to the
potential economic advantages of polyphosphoric acid (PPA) over the approved polymers,
refineries following market trends, or a combination of both. However, the performance of PPA-
modified asphalt binders is still uncertain. For instance, suppliers can use approved modifiers to
prepare a PG 76-22 from a PG 64-22, but they use PPA in a combination of styrene-butadiene-
styrene (SBS) or styrene-butadiene (SB) to reduce the stiffness of the final product. However,
PPA is a hydrophilic material; therefore PPA-modified asphalts may be susceptible to stripping.
Another concern of PPA-modified asphalt is how the modified binder interacts with different
liquid anti-stripping agents (LAAs) and various aggregates. Thus, it is very important to know
how PPA may affect rheological and performance properties of the base binder, but it is also
important to know whether PPA can be approved as an asphalt binder modifier for
environmental conditions and aggregates in Arkansas.
TRC 1501 Final Report
Page | 11
2. Background and Literature Review
To achieve the objectives of this study, first a comprehensive literature review was conducted.
The primary goal of doing a literature review was to gain knowledge about the methodology of
PPA modification and mechanistic performance of PPA-modified asphalts. To this end, pertinent
literature from well reputed journals and transportation agencies such as Transportation Research
Record, Journal of Materials in Civil Engineering, Federal Highway Administration (FHWA),
National Co-operative Highway Research Program (NCHRP), and Texas Transportation Institute
(TTI) were reviewed.
The term Useful Temperature Interval (UTI) is used to identify the difference between
the high and low temperature grading of an asphalt binder in the Superpave grading system. It
depends upon the chemical compositions of the binder. As a crude source can be used to produce
an asphalt binder of constant UTI, an asphalt producer desires to produce the asphalt binder with
the widest UTI range. Baumgardner (2009) stated that to achieve an UTI higher than 92°C (e.g.,
70+22=92 for a PG 70-22) neat binders must be modified by means of polymer(s) and/or
chemical modifier(s). It is also stated that binders with a narrow UTI (e.g., 86°C or 89°C) might
also require such modifications to meet the Superpave specifications. With PPA modification the
same crude source is capable of producing PG 64-22, PG 70-22, and PG 76-22 binders. The later
was affirmed by the researcher as a viable substitute of 3% SBS modified PG 64-22 asphalt
binder.
Masson. J. F. (2008) described PPA as an oligomer of Phosphoric Acid (H3PO4). The
basic compounds for the production of PPA are Phosphorus Pentoxide (P2O5) and Phosphoric
Acid (H3PO4). It is sold on the basis of its calculated content of H3PO4 as for example 115%.
Thus, PPA is available in various grades with percentages sometimes higher than 100% such as
105% or 115%. The 100 grade PPA (phosphoric acid) contains 72.4% P2O5, which is calculated
from the formula weight ratio P2O5/H3PO4. Similarly, Pyrophosphoric Acid (H4P2O7) contains
79.8% P2O5 as calculated from the ratio P2O5/H4P2O7. The ratio of P2O5 contents provides the
relative phosphoric acid content, which in this case is 79.8%/72.4% = 110%.
Pamplona et al. (2015) studied the effect of different proportions (0.0, 0.5, 1, 1.5, and
2%) of PPA on two 50/70 penetration grade asphalt binders of different chemical compositions.
The chemical changes induced by the addition of PPA were determined by thin-layer
TRC 1501 Final Report
Page | 12
chromatography and the researchers found that the effects of PPA were influenced by the
chemical composition of asphalt binder. It was reported that the effects of lower amounts of PPA
on chemical composition were more significant than those of higher amounts. In regard to the
percent recovery of the Multiple Stress Creep Recovery (MSCR) test, these researchers found
that PPA seemed to be a very good modifier and higher proportions of PPA (1.5 and 2.0%)
showed equivalent results. These researchers also reported that the effect of PPA decreased with
an increase in temperature. Finally, these researchers analyzed Linear Amplitude Sweep test
(LAS) test results to evaluate the fatigue characteristics of asphalt binders, and they suggested
that asphalt binders became more resistant to fatigue cracking with the addition of PPA under
both short- and long-term aging conditions.
Li et al. (2011) studied the effects of different modification types on the rheological and
damage properties of asphalt binders modified with PPA and different polymers. The base binder
of their study was PG 52-34, and was modified by 0.75% PPA, 0.3% PPA plus 1.0% styrene-
butadiene-styrene (SBS) polymer, 2.0% SBS polymer only, and 0.3% PPA plus 1.1% Elvaloy
polymer. Liquid phosphate ester was used as an anti-striping agent (ASA) at 0.5% (by the
weight) to all modified binders except for the SBS only modifier. These researchers used
Superpave DSR rheological tests for investigating high temperature PG grades of the modified
binders, and found that when the properties of recovered binders from field mixes were
considered, the PPA only and PPA plus SBS binder passed PG 70-XX, and SBS only and PPA
plus Elvaloy binders passed PG 64-XX. They also found that at lower temperatures long-term
aged recovered PPA plus Elvaloy and SBS only binders retained their -34oC grades, but the PPA
only and PPA plus SBS stiffened to have a low PG temperature of -28oC. From the MSCR test
results, the researchers found that the SBS plus PPA binder had the highest percent recovery
among the four modified binders. The authors also mentioned that the polymer plus PPA
modification was more resistant to fatigue cracking than the PPA only modification.
Kodrat et al. (2007) compared the effect of PPA-modified asphalt binders with straight
and polymer-modified binders. The researchers studied binders from 19 different crude sources,
and the PPA used to modify them contained 115% H3PO4. The tests performed by them to
analyze the effects of PPA in asphalt binders were Superpave performance grades, extended
bending beam rheometer, brittle state fracture and ductile state fracture. These investigators
found that depending on the crude source, the high temperature grade property of asphalt is
TRC 1501 Final Report
Page | 13
significantly increased by the addition of 0.75% PPA while the low temperature grading
properties remained largely unchanged. Moreover, the effect of PPA on the fracture properties in
the brittle state and on the reversible aging process was found to be insignificant. Finally, these
researchers recommended being cautious in using PPA-modified asphalt binders in areas where
pavements would be susceptible to fatigue cracking because they have found that PPA may
reduce the strain tolerance of asphalt in the ductile state.
Huang et al. (2008) studied the long-term aging characteristics of PPA-modified asphalt
binders from three different asphalt sources under un-aged and PAV-aged conditions. The
researchers used 1.5 % PPA of 105 grade (by the weight) to modify asphalt binders to carry out
DSR, Fourier transform infrared (FTIR) spectroscopy, differential scanning calorimetry (DSC),
and nuclear magnetic resonance (NMR) spectroscopy tests. The authors suggested that fatigue
cracking and low temperature cracking of the pavement should be reduced because of the
increment of the initial stiffness and low temperature flow properties of asphalts after long-term
aging by the addition of PPA. Moreover, based on the results of NMR spectroscopy, the
researchers were certain that the addition of PPA did not significantly change the internal
structure of the asphalt binders by chemical reaction even after long term aging. Furthermore,
from the NMR microscopy test results, the researchers affirmed that PPA reacts with residual
water present in asphalt and as little as 0.073% of residual water is sufficient to hydrolyze 1.5%
mass of PPA into orthophosphoric acid. Further, the researchers used a DSC device to evaluate
thermal properties of unmodified and PPA-modified asphalt binders under oxidative aging
conditions and reported that the locations, heights, and widths of the glass transition for PPA-
modified asphalts were parallel to those of the unmodified asphalts.
Jafari et al. (2015) studied the effects of stress levels on creep and recovery behavior of
modified asphalt binders. The researchers used one set of Rotational Thin Film Oven (RTFO)-
aged PG 58-XX binders modified with 2%, 4% and 6% SBS, and another set of RTFO-aged PG
58-XX binders modified with 0.5%, 1% and 1.5% PPA to carry out MSCR tests. The authors
stated that PPA-modified binders were less rutting resistant than the SBS-modified binders based
on higher non-recoverable compliance (Jnr) values of PPA-modified binders at 12.8 kPa than
SBS-modified binders. Moreover, they recommended adding a stress level of 12.8 kPa to the
MSCR standard procedure to compare nonlinearity and rutting resistance of PPA-modified
asphalt binders. Also, at high stress levels, PPA-modified binders did not exhibit their
TRC 1501 Final Report
Page | 14
viscoelastic properties, but the corresponding SBS-modified binders did, which convinced the
researchers to state that even at 1.5% of PPA concentration could result in overly sensitive
materials that would not be recommended for paving application. Finally, the authors
recommended that at a temperature higher than the performance grade of the modified binders,
the stress levels of lower than 3.2 kPa should be applied to avoid the negative recovery values of
the PPA-modified binders.
Huang et al. (2011) studied the rheological and chemical properties of hydrated lime and
PPA-modified asphalts. The researchers used 1.5% PPA and 10% hydrated lime (by weight) to
modify three SHRP core asphalt binders as the base binders. The rheological properties of un-
aged and PAV-aged binders were measured with a DSR at 30, 40, 58 and 64°C, and the chemical
properties were determined by using an NMR spectroscopy test. These researchers found that the
addition of PPA increased the rheological index substantially than the original neat asphalts.
Moreover, the addition of PPA increased the asphalt binder’s PG. However, further addition of
hydrated lime to the PPA-modified asphalt binder changed the PG back to a lower grade.
Furthermore, the researchers also found that addition of PPA increased the stiffness of the
asphalt binder, but further adding hydrated lime with it reduced this stiffness. Based on NMR
test results, the researchers reported that hydrated lime reacted with PPA in asphalt binder and
formed solid calcium phosphate, which also supported the rheological results obtained from the
DSR tests.
Baumgardner et al. (2005) described the mechanism of chemical modification of asphalt
binder with PPA. They analyzed two sets of modified and unmodified asphalts from two crude
sources, Saudi and Venezuelan, for chemical composition by asphaltene precipitation through
thin layer chromatography (TLC), nuclear magnetic resonance (NMR), gel-permeation
chromatography (GPC), and atomic force microscopy (AFM) tests. From the TLC test results,
the researchers found that the asphaltene content increased from 9.1 to 14.7 % (by weight) by
adding 1.2 % PPA to produce a PG 70-22 binder for the Saudi crude. On the other hand, for the
Venezuelan binder, the asphaltene content increased from 10.5 to 14.9% (by weight) with the
addition of 0.62% of PPA. The NMR analysis revealed that PPA preferentially reacted with the
asphaltenic phase of the asphalt. Finally, these researchers implied that the PPA changed the
chemical composition of the binder by changing saturates into asphaltenes.
TRC 1501 Final Report
Page | 15
Daranga et al. (2009) investigated the storage stability and effects of mineral surfaces on
PPA-modified asphalt binders. The researchers used three binders of varying grades to determine
their chemical compositions, while they used two unmodified binders of known composition
with two mineral fillers for determining the effect of mineral surface on asphalt binders. For the
study of storage stability, the researchers used PG 58-22, SBS-modified PG 70-22 and Elvaloy-
modified PG 64-22 binders further modified with 1% PPA (105 grade; by the weight of the
binder). On the other hand, for detecting the influence of asphaltenes on rheological properties of
PPA-modified asphalt binders, the researchers used PG 64-22 binder with 16% asphaltene
content, and PG 58-22 binder with 9% asphaltene content. The researchers found that for a given
binder, the higher the surface area to the volume ratio, the higher is the rate of oxidization.
Moreover, from observing the pattern of reaction with PPA molecule and hydrogen bond
formation, they concurred that the rate of oxidation of the modified asphalt binder did not depend
on the addition of PPA, rather for some binders, it slowed it down. Finally, the researchers stated
that the addition of mineral fillers increased the stiffness of the asphalt binder, whereas fillers
with basic pH showed a tempering effect on the stiffness increment of the modified binder.
Shulga et al. (2012) investigated the effect of foaming on the performance of PPA-
modified asphalt binders. The researchers used a Wirtgen WLB-10 laboratory foaming machine
to foam two sets of base binders, namely, PG 64-22 and PG 64-16, of different asphaltene
contents. PPA was used by the researchers to increase the PG of the base binders by one and two
grades. The actual amount of PPA used in their study was 0.8% by wt. for the one-grade increase
and 1.5% for the two-grade increase. The researchers performed Rotational viscosity, DSR,
MSCR, LAS and Binder Bind Strength tests to evaluate the performance of PPA-modified
foamed asphalt binders. They found that PPA did not have significant effects on the viscosity of
the foamed asphalt binders. The DSR test results revealed that PPA did not change the true grade
of the foamed binders. The researchers also found that PPA-modified foamed asphalt binders
showed similar or better bond strength than the neat binders, which means that the PPA might
have increased the bond between bitumen and aggregate. Finally, the LAS test results revealed
that the fatigue life of the foamed asphalt binders increased with the amount of PPA.
Maine DOT (2013) published a report on field performance of a PPA-modified asphalt
pavement. The field testing was performed on Route 1 in Perry where a pavement was
constructed using a 12.5 mm mix. The base binder of the project was a PPA-modified PG 64-28
TRC 1501 Final Report
Page | 16
binder, and the results were compared with a pavement constructed with a neat PG 58-28 binder.
The Automatic Road Analyzer (ARAN) data was used for evaluating the field performance of
the constructed pavement. It was reported that visual observation did not find any pavement
deterioration or distress in the PPA-modified pavement.
FHWA (2012) published a technical brief about the use and performance of PPA
modified asphalts. The agency, in cooperation with Transportation Research Board and
Minnesota DOT, organized a workshop to discuss using PPA as asphalt modifier. They found
that when 3% or higher dosage of PPA is necessary to increase the PG of asphalt to two grades,
the intermediate stiffness of the binder might increase. Also, the amount of PPA needed to
increase the grade of asphalt depends upon the crude source. The agency stated that the
extraction of binders modified with PPA could be hampered by stabilizing chemicals in the
extraction solvent due to the presence of acid scavenging chemicals. Moreover, this report
indicated that moisture damage potential of PPA-modified asphalt was very limited when the
dosage level was between 1 and 1.5%. Finally, the report indicated positive field performance of
PPA-modified asphalt binders in several states where the PPA dosage level was between 0.25
and 1.2%.
D’Angelo (2009) investigated the effects of PPA on asphalt binder properties from
different crude sources namely Saudi and Venezuelan. The researcher used a combination of
SBS polymer from different sources and PPA to modify PG 64-22 asphalt binders. This study
also used different blending temperatures and speed to blend the polymer and PPA to investigate
the difference in performance of the asphalt binders. It was found that the addition of 0.5% PPA
increased the high temperature PG grade of the Venezuelan binder to one grade, whereas for the
Saudi binder the same amount of PPA could not change the grade. Moreover, modified binders
blended at lower temperatures showed significantly lower MSCR % Recovery values in the
MSCR test than the modified binders blended at higher temperatures. Further, the addition of
PPA also made the MSCR % Recovery value higher than the SBS-modified binders. Finally, this
researcher used hydrated lime with PPA-modified asphalt to check for any negative interactions
with PPA. The researcher found no negative effects on stripping resistance, but the stiffening
effect of the PPA was found to be reduced to a small extent after the addition of lime.
TRC 1501 Final Report
Page | 17
Al-Qadi et al. (2014) investigated the effects of different additives and modifiers on the
moisture susceptibility of asphalt concrete. The researchers used LAA and hydrated lime as ASA
additives, and they used SBS and PPA as the modifiers. Four dosages (0.25, 0.5 0.75 and 1%) of
the LAA were used to optimize the LAA content in asphalt mixes. Moreover, four LAA
application methods (hydrated lime slurry, hydrated lime slurry with marination, dry hydrated
lime to dry aggregate and dry hydrated lime with moist aggregate) were used for selecting the
optimum hydrated dosage rate of hydrated lime. The researchers used 1.25% of PPA by weight
of asphalt binder to bump the grade of PG 64-22 to a PG 70-22 binder. Moreover, they used
1.5% SBS with PG 64-22 binder to bump the grade to a PG 70-22 binder. For determining the
moisture susceptibility they researchers performed Lottman test with five freezing and thawing
cycles, Hamburg wheel tracking test and fracture test using semi-circular bend (SCB). The
researchers also used SFE to characterize the adhesion and cohesion energy of asphalt binders
with various additives and modifiers. The SFE values showed that LAA had the highest work of
adhesion, and work of cohesion and compatibility ratio. However, the researchers found that
PPA reduced the rutting resistance compared to the control mixture, and the mixture with PPA
had the lowest fracture energy compared to the other modifiers. Finally, the researchers
concluded that the addition of PPA resulted in greater vulnerability to moisture damage
compared to the other additives and modifiers.
Yan et al. (2013) investigated the effects of PPA on chemical composition, physical
properties, and morphology of asphalt binder. The researchers used three different penetration
grade asphalts of 92, 85 and 63 to modify them with 105 grade PPA. The crude source of the
first two binders was Saudi Arabia, whereas the third binder was from Chinese crude. The
amount of PPA for modification was undisclosed in this study. The researchers performed
softening point, penetration and ductility test to characterize the physical properties of the
binders, whereas SARA analysis was done to characterize the chemical properties of the binders.
For morphological analysis, the researchers followed the atomic force microscopy (AFM)
technique. The researchers found that the physical performance and chemical composition of the
binders depend largely on the crude source. The binder with a lower colloidal index is greatly
influenced by PPA modification compared to the binder with a higher colloidal index. Finally,
the researchers concluded that with the addition of PPA, the asphaltene fraction of the base
binder increased and so did the viscosity of the modified asphalt binders.
TRC 1501 Final Report
Page | 18
Baldino et al. (2012) investigated the effects of PPA on low temperature rheological
properties on neat and PPA-modified asphalt binders. The researchers used four penetration
grade asphalt binders from three different crude sources: Venezuelan, Saudi Arabian, and
Russian. Moreover, they used three different dosages (0.5, 1 and 1.5% by the weight) of PPA of
115 grade for modification of the asphalt binders. For rheological analysis, the scientists used a
Dynamic Mechanical Analyzer (DMA) manufactured by Triton Technology, UK (TTDMA) for
dynamic mechanical analysis, and the ring ball softening test to determine the softening
temperature of the asphalt binder. The researchers found that the glass transition temperature of
asphalt binders decreases with the addition of PPA. Moreover, the complex modulus (E*)
increased with the addition of PPA, whereas the stiffness also increased with the addition of
PPA. Finally, the researchers stated that the effect of PPA on low temperature properties of
asphalt binder depended on the chemical composition and crude source.
Xiao et al. (2014) investigated the high temperature rheological properties of asphalt
binder modified four different polymer modifiers with or without PPA. Four polymers used in
this study were SBS, oxidized polyethylene, propylene-maleic anhydride and -40 mesh ambient
produced recycled crumb rubber. In this study, PPA was added with the polymer modified PG
70-22 binders. The PPA dosage rate was 0.5% by the weight of the asphalt binder. Two different
DOT approved sources of asphalt binders were used in this study. The researchers found that
using 0.5% PPA reduced the required polymer content to 1%. Moreover, the researchers found
that the failing temperatures were higher with PPA-modified binders and the rutting factors in
DSR test result were asphalt source dependent. Finally, the researchers concluded that the
general viscoelastic behavior of the modified asphalt binders was dependent on the type of
polymer used for modification.
Zhang et al. (2009) studied the influence of PPA, SBR, and sulfur on the physical
properties, rheological properties, morphologies and storage properties of asphalt binders. The
researchers used an AH-90 paving asphalt binder from China with a penetration grade of 90. The
PPA was blended with the base binder by using a high shear mixer at a speed of 5000 r/min. For
physical characterization, the researchers followed softening point test, penetration test, and for
chemical analysis they used FTIR test results. The researchers found that PPA improved the high
temperature physical and rheological properties. However, PPA had negative effects on the low
temperature ductility, but the addition of SBR improved the ductile properties. Finally, the
TRC 1501 Final Report
Page | 19
researchers stated that the addition of sulfur improved the high temperature rheological
properties of asphalt binder.
In short, as a summary to the literature review, existing literature suggests that PPA-
modified asphalt binders are less rutting resistant than corresponding polymer-modified binders.
But, fatigue resistance increases with the addition of PPA. Several researchers reported that the
change in the binder grade by PPA modification is crude source dependent. Further, a low
amount of PPA might yield favorable results and PPA modification gives better mechanistic
results if PPA is used in combination with another polymer modifier. Finally, from the findings
from the literature review and two surveys, the amounts of PPA used in this study were made.
TRC 1501 Final Report
Page | 20
3. Research Objectives
The principle objective of this study was to evaluate the short- and long-term laboratory
performance of PPA-modified asphalt binders, with a secondary investigation of asphalt
mixtures. Superpave binder tests were run on multiple asphalt binders and PPA dosage rates.
The laboratory performance of the asphalt binders also included an evaluation of the chemical
composition of the asphalt binders. Finally, selective asphalt mixture samples were prepared and
tested in the laboratory binders for their mechanistic performance and moisture resistance. Based
on the findings of this study, no negative effect of the selected levels of PPA was found in the
performances of asphalt binders. However, the dosage rates and performance characteristics of
the binder were found to be dependent on its crude source. Moreover, some admixtures such as
LAA were found to be incompatible with PPA-modified asphalt binders. Thus, ArDOT may
consider PPA as a potential modifier in additional to the existing co-polymers. However, it is
recommended that tests to confirm compatibility with the asphalt binder source and any additives
be performed before acceptance. This will require a revision of Subsection 404.01 Design of
Asphalt Mixtures (b) Design Requirements of Section 404 DESIGN AND QUALITY
CONTROL OF ASPHALT MIXTURES.
TRC 1501 Final Report
Page | 21
4. Research Methodologies
As discussed in Section 2, a comprehensive literature review was performed to gather
information from published scientific articles regarding PPA-modification. After this literature
review, two surveys were conducted to gain knowledge from two different perspectives, namely,
the users (transportation agencies) and the producers (refineries). The first survey was conducted
at the departments of transportation (DOTs), federal and other transportation agencies. The
second survey was conducted to the certified asphalt binder suppliers in Arkansas.
This chapter provides a summary of two surveys conducted to gather the most updated
knowledge regarding the state of the practice of the PPA modification. The first survey was
conducted at the departments of transportation, federal agencies etc. to gather information on any
guidelines, specifications, and concerns about using PPA to modify asphalt binder. The second
survey was conducted to the certified asphalt binder suppliers in Arkansas.
4.1 Survey Neighboring States
Methodology of State Survey: The survey was conducted using a web based survey platform
called “Survey Monkey,” in which ten questions were uploaded for conducting the survey before
sending out an e-mail invitation to the prospective respondents. Five polar (yes/no answer)
questions as well as five descriptive questions were selected for the survey. The survey was then
sent out to Transportation/Material specialists from departments of transportation, and federal
agencies etc. Afterwards, responses of all questions were collected automatically using “survey
monkey.” There are twenty seven respondents for this survey as of May 25, 2015. Figure 4.1
shows the state highway agencies that responded to this survey. The section below contains a
summary of the responses of 27 survey respondents. The total numbers of answers for a specific
question might be lower than 27, as some might have chosen to not answer any specific question.
Allowing PPA-modified Asphalt Binder in Construction Projects: Among the 27 respondents,
17 agencies allow PPA to modify asphalt binder, whereas eight of them do not allow PPA.
Figure 4.2 shows the respondents who answered this question. The green colored states answered
they allowed PPA as an asphalt binder modifier, whereas the red colored states did not allow
PPA.
TRC 1501 Final Report
Page | 22
Figure 4.1: Respondents of State Survey.
Figure 4.2: States Allowing PPA as a Modifier in Asphalt Binder.
Specifications of Using PPA for Binder Modification: Six DOTs of Ohio, Florida, Alabama,
South Carolina, Georgia and Mississippi have specifications of using PPA to modify asphalt
binders. The specification of Mississippi DOT states that PPA can be used at 0.75% to enhance
the physical properties of the base binder to meet the requirements for PG 67-22 binders.
Moreover, PPA may be used at a low dosage rate of 0.5% and as a catalyst or a mixing agent in
TRC 1501 Final Report
Page | 23
the production of performance grade PG 76-22 binders. The specification also strictly states that,
PPA can never be used to adjust the physical properties of the binder to a full binder grade. The
DOTs of Georgia, South Carolina, and Florida allow a maximum of 0.5% PPA per weight of the
binder to modify the neat asphalt binders, whereas the DOTs of Alabama and Ohio allow PPA in
the amounts of 0.2% and 1%, respectively. Figure 4.3 shows the states that have specification
about PPA modification among the respondents of this survey.
Figure 4.3: Existing Specifications Regarding PPA Modification.
Concerns of Using PPA-modified Asphalt Binders: A total of 16 DOTs expressed their
concerns of using PPA-modified asphalt binders, whereas ten DOTs said they were not
concerned about it. Figure 4.4 shows the graphical representation of the highway agencies’
concerns in this regard.
TRC 1501 Final Report
Page | 24
Figure 4.4: Concerns on Using PPA in Asphalt Binders.
Using PPA-modified Asphalt Binders in Last Five Years: Among the 27 respondents, 14
agencies used PPA-modified asphalt binders in their pavements, whereas eleven agencies said
they did not use any PPA-modified asphalt binders. Figure 4.5 shows the map of the respondents.
Figure 4.5: Using PPA-modified binders in Last Five Years.
TRC 1501 Final Report
Page | 25
Do Contractors Notify when They Use PPA-modified Asphalt Binders? Almost half of the
respondents said that the contractors do not notify them when they use PPA-modified asphalt
binders. Figure 4.6 shows the contractors’ notifying patterns regarding the use of PPA-modified
asphalt binders.
Figure 4.6: Response Whether Contractors Notify when PPA-modified Asphalt Binders is
Used.
Ongoing Research Projects on PPA-modified Asphalt Binders: A total of three DOTs,
namely, Maine, Oklahoma and New York, currently have ongoing projects to evaluate the
performance of PPA-modified asphalt binders. The name of the projects currently ongoing in
New York is “Determining Binder Flushing Causes in New York State.” The other two projects
currently being worked on in Maine and Oklahoma go by the names of “Field Test of a
Polyphosphoric Acid (PPA) Modified Asphalt Binder on Rt. 1 in Perry,” and “Performance of
Asphalt Binders Modified with Polyphosphoric Acid (PPA).”
Polymers Other Than PPA to Modify Asphalt Binders: From the responses collected in the
survey it is quite imminent that SBS is the most popular polymer modifier, which is used by
seventeen state highway agencies. The closest to SBS in terms of popularity amongst the twenty
seven respondents is SBR with ten users. Other polymers used by the DOTs to modify asphalt
binders are SB, GTR, Elvaloy, Crumbed Rubber (CRM) and aromatic oil. Figure 4.7 shows the
popularity of these polymers among the 27 respondents in this survey.
Yes
33%
No
52%
Skipped
15%
TRC 1501 Final Report
Page | 26
Figure 4.7: Popularity of Different Polymers Other Than PPA for Modifying Asphalt
Binders.
Incompatible Aggregate or Liquid Anti-stripping Agents (LAA): The only aggregate that the
respondents voiced their concerns while using them with PPA is limestone. Oklahoma DOT
stated that amine based LAA have the tendency to drop the true high temperature grade of the
asphalt binder, whereas phosphate based LAA drops the true grade less than the others.
Additional Information about PPA Modification: The respondent from the Utah Department
of Transportation said that the asphalt binder modification depended largely on the crude source,
and the refineries did not always notify them about modification so they do FTIR testing to
monitor any use of PPA or other polymers.
4.2 Survey Certified Asphalt Binder Suppliers in Arkansas
For the same purpose of gathering state of the art information regarding PPA modification of
asphalt binder, a separate survey was completed to certified asphalt binder suppliers in Arkansas.
The survey was sent to the asphalt binder suppliers listed in Table 4.1.
Methodology of Survey Certified Asphalt Binder Supplier in Arkansas: Ten questions were
finalized before sending the survey to the respondents. This survey was also conducted using the
web based survey platform called “Survey Monkey,” where the survey questions were uploaded.
An email invitation consisting of the URL of the survey and a message describing the purpose of
the survey was sent to each prospective survey participant. There were only three responses. The
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
SB SBS SBR GTR Elvaloy CRM Aromatic
Oil
Per
cen
tage
Polymer Type
TRC 1501 Final Report
Page | 27
following section in this chapter consists of the responses of the three respondents to the ten
questions.
Motivation of Using PPA: The first respondent stated that the combination of PPA with
polymer provides a cost effective mean of providing modified asphalt binder that meets the
needs of the public. However, the second respondent stated that they use PPA just to follow the
market.
Table 4.1: Certified Asphalt Binder Suppliers in Arkansas
Supplier Name Location
APAC-Central, Inc. Catoosa, OK
APAC-Missouri, Inc. Springfield, MO
Asphalt Terminals & Transportation, LLC. Muskogee, OK
Calumet Lubricants Company Shreveport, LA
Calumet Specialty Products Partners, LP Muskogee, OK
Coastal Energy Corporation Miller, MO
Coastal Energy Corporation Willow Springs, MO
Ergon Asphalt and Emulsion, Inc. Memphis, TN
Ergon Asphalt and Emulsion, Inc. Vicksburg, MS
Heartland Asphalt Materials Memphis, TN
Heartland Asphalt Materials New Madrid, MO
HollyFrontier Refining & Marketing LLC Catoosa, OK
Hunt Southland Refining Company Sandersville, MS
Hunt Southland Refining Company Vicksburg, MS
Lion Oil Company El Dorado, AR
Lion Oil Company Muskogee, OK
Marathon Petroleum Corporation Memphis, TN
NuStar Marketing LLC (NuStar Energy LP) Catoosa, OK
Phillips 66 Granite City, IL
Valero Marketing & Supply Company Ardmore, OK
TRC 1501 Final Report
Page | 28
Moreover, they also stated that they don't believe the final product quality and
performance using PPA matches that of elastomeric polymers and they will not use PPA at all if
possible. The third respondent answered that their refinery used PPA as a crosslinking agent
when modifying asphalt with Elvaloy polymers, and the PPA dosage is typically at a rate
between 0.15% and 0.30%. When used with SBS, PPA is used to reduce polymer concentration
for a given PG. This not only reduces cost but also improves rotational viscosity and makes some
highly modified binders easier to handle in the field. The third responder also stated that they
currently do not use PPA alone in modifying binders, but there are some cases where it would be
beneficial.
Mixing Protocol for PPA modification: As a response to the mixing protocol, the first
respondent said that they mix PPA after crosslinking near production temperature in line using a
static mixer. The final cure takes place in storage tank over a few hours. The second respondent
answered that they use a small batch tank where the asphalt is pumped (320 – 240oF) and the
PPA is added. The third respondent stated that PPA is added to modify binders at typical storage
temperatures (290 - 320oF). The mixing time is dictated by the size of the tank and the viscosity
of the base binder. In some instances, PPA is blended at the rack as the truck is loaded.
Preferable PPA Grade for Modification Purpose: The first respondent said they use 105 grade
PPA for asphalt modification purposes. The second respondent stated that to meet a PG 64-22V
(MSCR grading) specification they typically require 1.25% PPA. The third respondent answered
that they use both 105 and 115 grades of PPA, at dosage rates ranging from 0.15% to 1.0%.
Concerns from the Users Regarding PPA Modification: The first respondent said that the
users need to be knowledgeable and informed such as the concentration limit and verify mix
compliances. The second respondent answered that where anti-strips are present, the PPA is
removed from the system, and they are aware of a number of projects in different states that
failed due to the resulting asphalt being too soft after the acid or amine reactions take place.
Moreover, this respondent also stated that there are a number of states that do not allow the use
of PPA and many states limit the use of PPA. The third respondent stated they had not heard of
any concerns with the performance of PPA used in any projects, and the vast majority of binders
they supplied in some states included a combination of PPA and SBS in modified binders. The
respondent also mentioned that some of the states that they routinely supply binders completely
TRC 1501 Final Report
Page | 29
forbid the use of PPA. Furthermore, the respondent shared one of his/her experiences regarding
PPA modification where a customer added a non-PPA compatible warm mix additive to the
asphalt without the refinery’s knowledge. He/she also stated that certain warm mix additives will
negate the effects of the PPA and the reaction will also negate the effects of the warm mix
additive.
Other Additives Used with PPA: As an answer to this query, the first respondent mentioned
that they used SBS and crosslinker to obtain the desired elasticity. The second respondent stated
that they did not use any other modifier with PPA, whereas the third respondent answered that
they used SBS and Elvaloy with PPA.
Significance of Crude Source for PPA Modification: The first respondent stated that the
supplier must confirm the suitability when the source changes. The second respondent answered
that the asphalt (crude) source is critical to any formulation. For example, if the refinery
performs a caustic wash on the crude oil in their process, this would nullify the benefits of PPA
and effectively remove it from the system. Finally, the third respondent stated that asphalt
binders from certain crude sources cannot be modified well with PPA.
Additional Information Regarding PPA Modification: The first respondent stated that the
FHWA has done a considerable amount of work on PPA-modified asphalt binders, and they
should be consulted for additional technical information. The second respondent stated that he is
dubious about the long-term properties of liquid asphalt when PPA is used as opposed to
elastomers. The respondent prefers the MSCR specifications when a polymer curve is included,
which requires the use of an elastomer with or without the addition of PPA. Moreover, the
respondent’s opinion is that PPA meets the AASHTO specification, but its use compromises the
quality or performance. Thus, there are risk factors involved with some systems that can
compromise the quality of a project.
4.3 Test Plan
From these surveys, important information such as PPA dosage rates, compatible admixtures,
market culture, and information on aggregate compatibility were collected. From the findings of
the literature review, responses from survey participants, expertise of the industry partner
(Paragon Technical Service, Inc.), and in discussions with the ArDOT TRC research panel
TRC 1501 Final Report
Page | 30
members, a detail test plan of the study was developed. A summary of the research methodology
is presented in Figure 4.8. At first, the unmodified and modified asphalt binders were collected
from Paragon Technical Services, Inc. Selected PPA-modified asphalt binders were further
modified with LAA to evaluate their performance properties. The LAA were blended with hand-
blending technique previously implemented by the research group (Hossain et al. 2015)
Figure 4.8: Summary of Research Methodology.
Asphalt binders from two different sources were used to evaluate their mechanistic,
chemical and moisture susceptibility of the PPA- and LAA-modified binders. These properties
were compared with those of SBS-modified binders. For mixture performance tests, selected
types of binders from one source were used. The asphalt binder was a Canadian crude source,
and it was supplied by Ergon Asphalt and Emulsions, Inc. Memphis, TN. The second binder was
an Arabian crude source, which was a combination of “sweet and sour crudes,” and it was
supplied by Marathon Petroleum Corporation, Catlettsburg, KY. The second crude source was
selected purposely so that a different amount of PPA was needed to increase its PG grade from
PG 64-22 to PG 70-22. For each binder, three different dosages of PPA were used to modify the
TRC 1501 Final Report
Page | 31
neat binder. One of the PPA-modified binders from each binder source was foamed to evaluate
its performance as a Warm Mix Asphalt (WMA). Along with PPA, SBS was used as a modifier
to compare properties of PPA- and SBS-modified binders. Details of sample modifications and
nomenclatures are shown in Table 4.2. As shown in this table, a known compatible LAA named
Kao Gripper® X2 (0.5% by weight; samples S1B4 and S2B4) from Kao Specialties Americas,
LLC was included in the test plan, as suggested by Paragon’s Chemist. All binder samples
except the foamed binders presented in Table 4.2 were blended in Paragon’s laboratory. The
foamed asphalt binders were prepared by using “The Foamer” available at the ArDOT Materials
Laboratory.
Table 4.2: Details of Sample Nomenclature
Base
Binder
Crude
Source Refinery Modification
Final
Performance
Grade
Sample
Nomenclature
PG 64-22
Canadian
Ergon
Asphalt &
Emulsions,
Inc.,
Memphis,
TN
- PG 64-22 S1B1
0.25% PPA Softer than
PG 70-22
S1B2
0.5% PPA PG 70-22 S1B3
0.5%PPA, 0.5%
LAA PG 70-22
S1B4
0.5% PPA (Foamed) PG 70-22 S1B5
0.75% PPA Harder than
PG 70-22
S1B6
2% SBS PG 70-22 S1B7
2% SBS, 0.5% PPA PG 76-22 S1B8
Arabian
Marathon
Petroleum
Corporation,
Catlettsburg,
KY
- PG 64-22 S2B1
0.5% PPA Softer than
PG 70-22
S2B2
0.75% PPA PG 70-22 S2B3
0.75%PPA, 0.5%
LAA PG 70-22
S2B4
0.75% PPA
(Foamed) PG 70-22
S2B5
1% PPA Harder than
PG 70-22
S2B6
2% SBS PG 70-22 S2B7
2% SBS, 0.75% PPA PG 76-22 S2B8
TRC 1501 Final Report
Page | 32
From the findings of literature review and survey results, it was also determined that some
highway agencies were concerned about the performance of several LAAs present in the market.
Since PPA is an acid, a possible compatibility issue between PPA and LAA develops. At present,
ArDOT allows LAAs from four particular suppliers, namely, Akzo Nobel Surface Chemistry
LLC, Arr-Maz Custom Chemical, MeadWestvaco (MWV) Specialty Chemicals (currently called
Ingevity), and PreTech Industries, Inc. LAAs produced by these suppliers were used to further
modify (hand blend) PPA-modified PG 70-22 binders and tested in the A-State laboratory. As
recommended by the manufacturers, the following dosage levels were used in this study: (i)
0.5% AD-here® HP Plus™ from Akzo Nobel, (ii) 0.5% PermaTac Plus
® from Arr-Maz, (iii)
0.5% Evotherm®
M1 from Ingevity, and (iv) 0.5% PaveGrip® from PreTech. The nomenclatures
used for the LAA modified samples are shown in Table 4.3.
Table 4.3: Nomenclatures of LAA Modified Binders
Base Binder LAA Nomenclature
S1B3
Pavegrip S1B4-Pavegrip
PermaTac Plus S1B4-PermaTac
Adhere HP Plus S1B4-Adhere
Evotherm M1 S1B4-Evotherm
S2B3
Pavegrip S2B4-Pavegrip
PermaTac Plus S2B4-PermaTac
Adhere HP Plus S2B4-Adhere
Evotherm M1 S2B4-Evotherm
S1B1
Pavegrip S1B1+Pavegrip
PermaTac Plus S1B1+PermaTac
Adhere HP Plus S1B1+Adhere
Evotherm M1 S1B1+Evotherm
S2B1
Pavegrip S2B1+Pavegrip
PermaTac Plus S2B1+PermaTac
Adhere HP Plus S2B1+Adhere
Evotherm M1 S2B1+Evotherm
TRC 1501 Final Report
Page | 33
4.4 Test Methods
4.4.1 Asphalt Binder Rheological Tests
Penetration Test
Penetration test of asphalt binder is the measurement of the depth of penetration of a standard
needle into the asphalt binder. The test is performed in accordance with AASHTO T 49. The test
procedure includes melting and cooling the asphalt binder to room temperature. A standard
penetration needle is allowed to penetrate into the binder for a period of 5 seconds. The depth of
penetration is measured in units of 0.1 mm and recorded as penetration number. The total weight
allowed to penetrate into the asphalt binder is 100g. Figure 4.9 shows a penetration test device.
Figure 4.9: Penetration Test Device.
Rotational Viscosity (RV) Test
The RV test was performed in accordance with AASHTO T 316. Figure 4.10 shows a DV-II+
Pro rotational viscometer (RV) from Brookfield Engineering Inc. in which the test was
TRC 1501 Final Report
Page | 34
performed. The RV test is performed to measure the viscosity of asphalt binders at higher
temperatures. In this study RV test was done from 135 to 180 °C at a 15° C interval.
Figure 4.10: RV Test Device.
The viscosity of asphalt binder is the measure of the workability, pumpability, and
mixability of the asphalt binder. At first, the asphalt binder sample is heated until fluid and 10
gm of asphalt binder is poured into the sample chamber. The temperature is set to the desired
temperature by using a temperature controller and it is kept for 30 minutes to bring it to the set
temperature. At that temperature, the motor is turned on to rotate the spindle at a constant speed
of 20 RPM. The amount of torque required maintaining the constant speed (20 RPM) of the
cylindrical spindle is used to estimate the viscosity of the binder. After 10 minutes of
temperature equilibrium, 3 separate readings are taken at 1 min interval. The Superpave
specification for unaged asphalt binder is that the viscosity of the binder should be ≤ 3 Pa.s at
135°C.
Dynamic Shear Rheometer (DSR) Test
The dynamic shear rheometer (DSR) test is performed to characterize the viscous and elastic
behavior of asphalt binder at high and intermediate service temperatures. The DSR measures the
TRC 1501 Final Report
Page | 35
complex shear modulus (G*) and phase angle (δ) of asphalt binders at desired temperatures and
frequency of loading. The G* is the measure of the total resistance of the binder to deformation
when repeatedly sheared whereas, the δ is the measure of elasticity of the binder. The lower the
values of δ, the more elastic the binder is, whereas a higher value indicates viscous binder.
Figure 4.11 shows an Anton Paar MCR 302 DSR machine which was used in this study.
In the DSR test, a thin binder sample is sandwiched between two circular plates where the lower
plate is fixed and the upper plate oscillates back and forth at a certain frequency, creating a
shearing action. According to AASHTO T 315, the test frequency is 10 radians per second (1.59
Hz). The test is performed according to AASHTO T 315 in different aging conditions, namely,
unaged, RTFO-aged and PAV-aged, of the binders. For unaged and RTFO-aged binders, the
primary measurement according to the Superpave specification is the rutting parameter, which is
calculated by taking the ratio of G* and sinδ (i.e., G*/sinδ). On the other hand, the DSR test for
PAV-aged binders calculates fatigue factor at intermediate temperatures by multiplying G* and
sinδ (i.e., G*sinδ).
Figure 4.11: Dynamic Shear Rheometer.
TRC 1501 Final Report
Page | 36
The Superpave specifications with respect to the DSR test results for unaged, RTFO-aged and
PAV-aged binders are shown in Table 4.4.
Table 4.4: Superpave Specification for Rutting and Fatigue Factor
Material Value Test Temperature (oC) Specification
Unaged binder G*/sinδ High Service ≥ 1.0 kPa (0.145 psi)
RTFO-aged binder G*/sinδ High Service ≥ 2.2 kPa (0.319 psi)
PAV-aged binder G*sinδ Intermediate Service ≤ 5000 kPa (725 psi)
Bending Beam Rheometer (BBR) Test
The BBR test is performed to measure low temperature stiffness and stress relaxation properties
of asphalt binders. These parameters indicate asphalt binders’ resistance to low temperature
cracking. Apart from that, BBR test also provides the low service temperature of the PG grading.
From the BBR test, creep stiffness and the slope of the master stiffness curve, referred to as “m-
value”, at 60 s is measured. The test is performed in accordance with AASHTO T 313. A typical
BBR device is shown in Figure 4.12. The Superpave specifications for BBR test are shown in
Table 4.5.
For the test, degassed PAV-aged binders are used to prepare a 0.246 x 0.492 x 5.000 inch
(6.25 x 12.5 x 127 mm) solid asphalt beam. This beam is loaded at its midpoint in a simply
supported set-up where the two supports are 4.02 inches (102 mm) apart and the load is 0.22 lb
(100 g). The beam deflection is measured at 8, 15, 30, 60, 120 and 240 seconds. A stiffness
master curve is plotted for these points. From the curve, slopes are drawn at 8, 15, 30, 60, 120
and 240 seconds to calculate the “m” values. The test is performed at a 10°C higher than the
expected the low service temperature. To simulate the low service temperature, the time-
temperature superposition principle is used.
TRC 1501 Final Report
Page | 37
Figure 4.12: Bending Beam Rheometer (BBR).
Table 4.5: Superpave Specification for BBR Test
Parameter Test Temperature (oC) Specification
“m-value” at 60 second Low Service Temperature +10oC ≥ 0.300
Stiffness at 60 seconds Low Service Temperature +10oC ≤ 300 MPa
Rotational Thin Film Oven (RTFO)
The RTFO oven simulates short term aging of asphalt binders for use in DSR test as well as for
PAV-aging. The RTFO oven uses high temperature and air pressure to simulate the aging
phenomenon that happens to asphalt binders during the heating and storage inside of a mixing
plant. Figure 4.13 shows an RTFO oven used for this study. The RTFO-aging of asphalt binders
is done according to AASHTO T 240. At first, 35 gm asphalt binder is poured into cleaned and
TRC 1501 Final Report
Page | 38
preheated RTFO glass bottles. The glass bottles are then placed into the RTFO sample rack
which rotates at a speed of 15 rpm. The test temperature is 163°C and the aging time is 85
minutes. During the test, 244 in3/min (4 L/min) air flows into each sample bottles.
Figure 4.13: Rotational Thin Film Oven (RTFO).
Pressure Aging Vessel (PAV)
The PAV simulates long term aging of asphalt binders (7 to 10 year period). The PAV aging is
done in accordance with AASHTO R 28. Figure 4.14 shows the PAV device used for this study.
The aging process is conducted at various temperatures namely, 90, 100 and 110°C depending
on the climatic condition. For this study a 100°C aging temperature was selected. Moreover, the
aging process takes 20 hours. The required air pressure for PAV aging is 300 psi (2.07 MPa).The
PAV-residues are used for DSR tests for measuring the fatigue factor and BBR test to measure
the low temperature cracking properties of asphalt binder. However, for using the PAV residues
for direct tension test (DTT), it is recommended to degas the sample in a vacuum degassing
oven. Other than DTT, the degassing activity is optional. Figure 4.15 shows a vacuum degassing
oven used in this study. The degassing process is done at temperature of 170°C for a period 30
minutes.
TRC 1501 Final Report
Page | 39
Figure 4.14: Pressure Aging Vessel (PAV).
Figure 4.15: Vacuum Degassing Oven.
TRC 1501 Final Report
Page | 40
4.4.2 Chemical Performance Tests
FTIR Spectroscopy
FTIR analysis is a common and quick technique to identify the functional groups present in
asphalt binders. It is commonly used in the asphalt industry to identify the presence of any
specific functional group in asphalt binders (Yildrim, 2007; Masson et al., 2001; Diefendefer,
2006; Fernandez-Berridi et al., 2006). The principle behind FTIR spectrum analysis could be
explained by the Planck-Einstein relation (Equation 4.1).
E = h ν = h c/λ = h c ΰ 4.1
In Equation 4.1, E is the energy, h is Planck’s constant, c is the speed of light and ΰ is the
wavenumber, which is inversely proportional to the wavelength. According to this principle,
when a given molecule absorbs energy a range of mechanical motion including symmetrical and
asymmetrical stretching, rocking, wagging scissoring and twisting is possible when photons with
discrete energy are present. This enables the mapping of the absorption bands in the FTIR
spectrum. Figure 4.16 shows the stretching and bending vibrations of atoms due to absorption of
infrared radiation. If one considers the incident infrared radiation intensity as “Io” and the
intensity of the beam after the interaction of the sample as “I”, the ratio “I” and “Io” is a function
of the frequency of light, which gives the spectrum. This spectrum could be specifically three
types, namely transmittance, reflectance and absorbance. The multiple vibration type occurring
instantaneously creates a very complex absorption spectrum. This picture of spectrum is
dependent on the functional group present in the sample which changes the value of “I” after the
interaction of infrared radiation. The FTIR device detects this intensity of light with the help of a
detractor after the infrared interact with the sample. A working principle of FTIR is shown in
Figure 4.17. For this study, a Nicolet 8700 spectrometer was used, and the FTIR spectrum was
analyzed using Thermo Electron’s OMNIC software.
TRC 1501 Final Report
Page | 41
Figure 4.16: Stretching and Bending Vibrations of Atoms due to Absorption of Infrared
Radiation.
Figure 4.17: Experimental Set-up for FTIR Spectroscopy.
Sample preparation is a very important step of FTIR test, as improper sample preparation
could change or alter the spectrum (Nasrazadani et al., 2010). There are various methods of
TRC 1501 Final Report
Page | 42
sample preparation of asphalt binder testing. One of the methods is the Attenuated Total
Reflectance (ATR) method and the other one is transmission FTIR method. The former method
is suggested by AASHTO T 302-05 to quantify the polymer content in asphalt binders. In this
method, 10 gm of heated asphalt binder is placed over a wax paper which is cut to the size
slightly larger than the face of an ATR crystal. Care has to be taken so that sufficient material is
placed on the crystal so that the entire face is covered. The desired thickness of asphalt binder
over the wax paper is 1 mm. Before running the test, the paper with asphalt binder is allowed to
cool down for a sufficient period of time.
In the transmission FTIR method, 10 gm of heated asphalt binder is placed on wax paper.
Over the wax paper, approximately 100 mg of potassium bromide (KBr) is pressed in a 13 mm
die of about 10000 psi for two minutes to achieve a solid KBr pallet.
Apart from the aforementioned sample preparation techniques disposable Real Crystal IR
cards is also used by researchers for FTIR sample preparation technique. This is a very recent
technique to accommodate samples such as asphalt binders. The IR card contains a circular
aperture which enables transmitting IR microporous substrate. For preparing the samples, at first
asphalt binder is heated to 163 °C to make it fluid. One drop of the fluid binder is placed on the
side of the IR card crystal aperture, which is spread over the crystal by a clean glass slide. The
sample must be thin enough for the source to pass through. After preparing the sample, it is
allowed to cool down sufficiently before running any test. For this study, KBr IR cards of 9.5
mm aperture size from International Crystal Labs were used. Figure 4.18 shows an empty IR
card and a prepared sample for FTIR test.
For this study, a KBr beam splitter from a spectrum range of 350 to 7400 cm-1
was used.
The samples were run over 50 scans at 4 cm-1
resolution for 30 seconds. The test was done at a
relative humidity under 5%. At first, the IR spectrum was taken of an empty IR card for
background, over which the samples were tested. The data analysis was performed by the
OMNIC software, which provides the absorbance and wavenumber data for a sample. The data
was plotted with the help of the MS Excel tool.
TRC 1501 Final Report
Page | 43
Figure 4.18: An Empty IR card (Left) and a Sample Ready for FTIR Test (Right).
Acid Detection Test
To detect the presence of phosphorus or PPA in asphalt binder, a qualitative test titled acid
detection test was done. The test result shows the presence or absence of PPA by a visible color
change in the reagents used for the test. The test is performed in accordance with AASHTO TP
78. The reagents needed for this test are ammonium molybdate, antimonyl tartrate, 1 n sulfuric
acid, ascorbic acid and butyl alcohol. Typical positive and negative results of acid detection test
is shown in Figure 4.19 (Figure 19a is a negative test, and Figures 4.19b and 4.19c are positive
test).
TRC 1501 Final Report
Page | 44
Figure 4.19: Acid Detection Test Results.
SARA Analysis
The SARA analysis was intended for determining the percentages of certain families of chemical
constituents in the tested asphalt binders. The SARA analysis was performed in accordance with
"ASTM D 4124-09: Standard Test Method of Separating Asphalt into Four Fractions". In this
method, the test specimen was put into reflux with n-heptane for at least three hours. To start a
reflux, an asphalt specimen weighing 2.00±0.30g was taken in a round bottom flask. For each
gram of asphalt specimen, 100mL of n-heptane (HPLC grade) was added to it. A stirring magnet
was put into the flask. A Leibig condenser was fitted to the opening of that flask. The assembly
was fastened with a clamp and set on a heating bath containing sand smaller than the size of the
US standard Sieve No. 20. The heating bath was then placed on a hot plate and the temperature
was set at 200±50°C and the stirring was set at 300±50 rpm. This reflux operation caused the
highly polar fractions (i.e. asphaltenes) to precipitate. Although the standard, ASTM D 4124-09,
recommended using iso-Octane, it was not capable enough to entirely dissolve the specimen in
reflux. Therefore, n-heptane was used to get the entirely dissolved specimen. The other three
constituents (saturates, aromatics, and resins) got dissolved in n-heptane. The mixture of those
three fractions is typically termed as maltenes. Maltenes were loaded onto a chromatographic
column of activated alumina (pH 9-10) of particle size 50-200 m and allowed to elute under
gravity. The individual fraction came out in a sequence as saturates, aromatics and resins. The
non-polar saturates came out first with n-heptane elution. The naphthene aromatics fraction was
eluted with a consecutive application of toluene and toluene:methanol (50:50) solvents. A UV
light of 366 nm wavelength was shined onto the column to monitor the advancement of the
TRC 1501 Final Report
Page | 45
aromatics fraction. A fluorescent band was progressing down. After collecting the entire
fluorescent band, the polar aromatics or the resins started to elute. The resins fraction was
completely collected lastly with trichloroethylene. All the eluted fractions were completely dried
using a rotary evaporator and were reported as the percent fraction of the original sample.
Sometimes a drying with chloroform was required to escape all the solvents out of the eluted
fractions.
Moisture Susceptibility Measurement by Surface Free Energy (SFE) Technique
Apart from the mixture tests, moisture susceptibility can also be measured by the surface
science-based SFE techniques. In the SFE method, cohesive energy of asphalt binder is
calculated and the adhesive forces between asphalt binders and aggregates are calculated.
Researchers have used the SFE technique to measure the compatibility between the asphalt
binder and aggregates to quantify the vulnerability of the asphalt-aggregate compatibility to
moisture damage. In this study, the Sessile Drop (SD) method has been used as it is simple, less
time consuming and involves easy sample preparation techniques. In the SD method, static
contact angles of asphalt binders and aggregates are measured by an Optical Contact Analyzer
(OCA) device. Figure 4.20 shows the OCA device used in this study.
Figure 4.20: Optical Contact Analyzer (OCA) Device.
TRC 1501 Final Report
Page | 46
In the SD technique, one drop of probe liquid is dropped on the asphalt binder sample
which is coated over a thin glass slide and subsequently the shape of the drop is automatically
analyzed by the software connected to the OCA device. For one drop, more than 100 contact
angles on each side of the drop are measured in this technique to get a very precise measurement.
The volume of the drop is regulated and the same drop volume is used for all the samples. To
determine the contact angle of asphalt binder a very smooth and thin surface of asphalt is created
on a thin glass slide. The sample preparation procedure for a SD test is described below:
At first, asphalt binders are heated at a temperature of 163 °C until they are fluid enough
to spread over a solid surface.
A glass slide of 57 mm x 70 mm x 1.5 mm is wrapped on all four sides with scotch tape
to the desired sample outline. The glass slide is cleansed by quickly run through a flame
before coating with the tape to make sure it is free from any particles or charges.
With the help of a trimmer, a very small amount of asphalt binder is placed over one of
the sides of the taped area.
Another glass slide is quickly pressed and moved starting from the asphalt drop over the
taped area to spread the asphalt binder evenly and smoothly over the surface of the glass
slide.
The tapes are removed after the asphalt binder has been cooled down sufficiently.
Figure 4.21 shows an asphalt binder sample ready for an OCA test. Beside asphalt binder
samples, contact angles of aggregate samples were also measured using the SD method for this
study.
TRC 1501 Final Report
Page | 47
Figure 4.21: An Asphalt Binder Sample Ready for OCA Test.
The sample preparation technique for an aggregate sample is briefly described below:
Cutting of the rock samples: Rock samples are cut to thin and acceptable sizes before
doing any types of cleaning or polishing. The rocks are cut in thin slices where the cut
thickness varying from 0.25 to 0.5 inches. Figure 4.22 shows a typical rock sample
after cutting.
Figure 4.22: A Typical Rock Sample After Cutting.
TRC 1501 Final Report
Page | 48
Cleaning and polishing of the sample: After cutting, the samples are washed with
water to remove any visible material present on the surface and cleansed using a
paper towel. After cleaning, the testing surface of the sample is polished by silicon
carbide grits. When the polishing is finished, the samples are washed with soap and
warm distilled water. After this initial cleaning, the samples are cleansed further by
hexane, which is applied to the sample by a saturated paper towel.
Drying: After cleaning, the samples are dried at 105 °C for 12 hours. Then the sample
is cooled to room temperature in a desiccator with anhydrous calcium sulfate crystal.
4.4.3 Asphalt Mixture Tests
Evaluation of Rutting and Stripping of Asphalt (ERSA) Test
The ERSA Test is used for testing the rutting and moisture-susceptibility of hot mix asphalt
(HMA). It determines the susceptibility of premature failure of HMA caused by weakness in the
aggregate structure, inadequate binder stiffness, or moisture damage. This test method aims to
measure the rut depth and the number of passes to failure. In order to perform this test, a saw-cut
slab from laboratory-compacted HMA specimen or a core from a compacted pavement is
needed. The thickness of the cylindrical specimen can be from 38 mm (1.5 in) to 100 mm (4 in)
and the diameter of 150 mm (6 in). For each test, two samples are needed, which are shown in
Figure 4.23. The specimen has to be submerged in a water bath of 40oC to 50
oC. The ERSA
Tracking machine has a moving wheel of 203.2 mm (8 in) and 47 mm (1.85 in) wide steel which
goes along the specimen. From a specification standpoint, it is essentially identical to the
Hamburg Wheel Test. The wheel has a load of 158 lbs (703 N), and it should make 52 passes
across the specimen per minute with a maximum speed of 0.305 m/s. The machine is limited to
running 20,000 cycles. Arkansas specifications (Section 407) for surface courses require a
maximum rut depth of 8.000 mm (0.314 inches) at 8,000 cycles for an APA style wheel tracking
tests. Since the University of Arkansas is utilizing the ERSA tester, the maximum cycle value of
8,000 cycles and a maximum rut depth of 8.0 mm (0.324 inches) are utilized. The gradation of
the specimen, which is caused by the wheel, is measured.
TRC 1501 Final Report
Page | 49
Figure 4.23: Left and Right ERSA Samples to Perform Test.
Dynamic Modulus in IDT configuration
IDT Dynamic Modulus is a modification of the standard for measuring dynamic modulus
according to AASHTO T 342. While AASHTO standard uses cylindrical specimens, IDT
dynamic modulus uses disc specimens, which are the same in shape as the ones used in creep
compliance tests (AASHTO T 322). The basis for using this configuration as a tool for
measuring IDT dynamic modulus was proposed by Kim et al. (2004) based on the mathematical
formulation developed by Hondros (1959). The test basically consists of applying loads at
varying frequencies (from 25 Hz to 0.1 Hz) at different temperatures (from -10oC to 54
oC). A
basic test setup is shown in Figure 4.24.
TRC 1501 Final Report
Page | 50
Figure 4.24: Dynamic Modulus in IDT Configuration.
Uniaxial Dynamic Modulus
In order to measure the Dynamic Modulus of an asphalt concrete sample, a dynamic load is
applied at different frequencies: 0.1 Hz, 0.5 Hz, 1.0 Hz, 5 Hz, 10 Hz, and 25 Hz. Each of these
frequencies as tested at five different temperatures: -10 , 4 , 21 , 37 , and 54 . The
amplitude of the load is recorded as well as the vertical deformation of the specimen using three
extensometers. Then, the ratio of the amplitudes of vertical stress to vertical strains is computed
in order to obtain the value of dynamic modulus for each combination of frequency and
temperature following AASHTO T 62-07 (AASHTO, 2014). Specimens of 100 mm (4 in.)
diameter and 150 mm (6 in.) tall were used, as seen in Figure 4.25.
Semi-Circular Bend (SCB) Fracture Test
Semi-Circular Bend (SCB) Test determines the fracture energy and fracture toughness at low
temperatures of asphalt mixtures by its semi-circular geometry. The geometry is a half disc with
a notch, as seen in Figure 4.26. To perform the test two samples are needed per test. Three
replicates per binder were prepared. From each of these, a slice from the center of the specimen
was obtained. This slice, which should be 25 mm (1 in.) thick, 150 mm (6 in.) diameter, and
TRC 1501 Final Report
Page | 51
notch’s length 15 mm (0.6 in.) and 1.5 mm (0.06 in.) width, is cut in a half where one half is
tested at -24oC and the other half at -10
oC. The loading rate applied was 0.03 mm/min (0.012
in/min), as per AASHTO TP 105-13 (AASHTO, 2013).
Figure 4.25: Uniaxial Dynamic Modulus.
Figure 4.26: SCB Test.
IDT Creep Compliance and Indirect Tensile Strength
Creep compliance is defined as the time-dependent strain divided by the applied stress and
indirect tensile strength is the strength shown by a specimen, which is subject to tension. The test
TRC 1501 Final Report
Page | 52
method will follow AASHTO T 322 specifications and will be conducted at the following
temperatures: -10, 0, and 10oC. Three replicates will be used and the specimen geometries will
have a diameter of 6 in. (150 mm) and a 1.5 in. height (38 mm). The purpose of the test is to
determine the tensile creep by the application of a static load of fixed magnitude along the
diametrical axis of the specimen for the duration of 100 seconds. During the creep test, loads are
selected to remain horizontal strain in the linear viscoelastic range, which is typically below 500
x 10-6
mm/mm. When the creep tests are finished at each temperature the tensile strength can be
determined by applying a load to the specimen at a rate of 12.5mm of ram movement per minute.
The vertical and horizontal deformation will be recorded on both ends of the specimens and the
load until the load starts to decrease.
Checking Mix Design
In order to perform the mix design a binder of PG 64-22 was used combined with aggregates ½”
Chips (11%), 3/8” VB Gr Chips (17%), Manufactured Sand (20%), 37% of ¼” Screening (37%),
and Concrete Sand (15%). Mix design HM154-11 was used from APAC-Central (Arkhola) in
Van Buren, AR.
Specific Gravity of Coarse Aggregate – AASHTO T 85
To perform the specific gravity of coarse aggregate, an amount of 2000 g coarse aggregate was
used. Table 4.6 shows the results obtained for specific gravity for each aggregate.
Table 4.6: Specific Gravity of Coarse Aggregate
Ag
gre
gate
Ma
ss, O
ven
dry
sp
ecim
en,
A
Ma
ss,
satu
rate
d
surf
ace
dry
spec
imen
, B
Ma
ss,
satu
rate
d
sam
ple
in
wa
ter, B
Sp
. G
rav
ity
Sp
ecif
ic
Gra
vit
y S
SD
Ap
pa
ren
t
Sp
ecif
ic
Gra
vit
y
Ab
sorp
tio
n
½” Chips 1980.4 2029.1 1228.9 2.47 2.54 2.64 2.46%
3/8” VB Gr Chips 1980.3 2043.8 1228.2 2.43 2.51 2.63 3.21%
¼” Screening 1966.3 2032.6 1222.4 2.43 2.51 2.64 3.37%
TRC 1501 Final Report
Page | 53
Specific Gravity of Fine Aggregate – AASHTO T 84
To perform the specific gravity of fine aggregate, an amount of 1000 g of fine aggregate was
used. Table 4.7 shows the results obtained for the specific gravity for each aggregate.
Table 4.7: Specific Gravity of Fine Aggregate
Ag
gre
gate
Ma
ss, O
ven
dry
spec
imen
, A
Ma
ss,
Py
cnom
eter
+
wa
ter,
B
Ma
ss,
py
cno
met
er +
Sp
ecim
en +
Wa
ter,
C
Ma
ss,
SS
D
spec
imen
, S
Sp
. G
r.
Sp
. G
r. S
SD
Ap
pa
ren
t S
p.
Gr.
Ab
sorp
tio
n
½” Chips 496.6 1273.6 1582.7 500.1 2.60 2.62 2.65 0.70%
3/8” VB Gr
Chips 495.5 1263.6 1568.9 500.1 2.54 2.57 2.61 0.93%
Man Sand 496.3 1266.2 1576.4 500.0 2.61 2.63 2.67 0.75%
¼” Screening 496.2 1253.8 1561.7 500.0 2.58 2.60 2.64 0.77%
Concrete
Sand 499.5 1272.1 1585.3 500.0 2.67 2.68 2.68 0.10%
*These data were collected from the fine fraction of the coarse aggregate.
Bulk Specific Gravity (Gmb) – AASHTO T 166
Three replicates of 4500 g were used, which were compacted with 75 gyrations combined with
amount of binder was 297.44 g. Three additional replicates of 4500 g were used, which were
compacted with 115 gyrations. Tables 4.8 through 4.10 show the results for the bulk specific
gravity.
Table 4.8: Bulk Specific Gravity (Gmb) at Ndesign of 75
Sample A (Mass Dry, g) B (Mass Sat. Air, g) C (Mass in Water, g) Gmb (A/(B-C) Va
Replicate1 4807.10 4809.20 2743.40 2.327 2.43
Replicate2 4732.40 4733.70 2702.80 2.330 2.29
Replicate3 4732.30 4734.10 2704.10 2.331 2.25
Average 4757.27
2.329 2.33
TRC 1501 Final Report
Page | 54
Table 4.9: Bulk Specific Gravity (Gmb) at Nmax of 115
Sample A (Mass
Dry, g) B (Mass Sat. Air, g) C (Mass in Water, g) Gmb (A/(B-C) Va
%G
mm
Replicate1 4837.10 4838.20 2770.00 2.339 1.93 0.98
Replicate2 4886.50 4887.30 2793.20 2.333 2.16 0.98
Replicate3 4628.20 4629.80 2641.00 2.327 2.42 0.98
Average 4783.93
2.333 2.17 0.98
Bulk Specific Gravity (Gmm) – AASHTO T 209
Three additional replicates of 4500 g were used combined with an amount of binder 297.44 g.
Table 8 shows the results of the bulk specific gravity (Gmm).
Table 4.10: Bulk Specific Gravity (Gmm)
Sample A (Mass Dry, g) C (Mass in Water, g) Gmm (A/(A-C)
Replicate1 2792 1623.4 2.389
Replicate2 2642.5 1532.5 2.381
Average 2717.25
2.385
4.5 Field Performance Data Collection
It is imperative to collect field performance data such as roughness and rutting to check the
current condition of the road network, the deterioration rate, or the necessity for maintenance by
a highway agency. Almost all North American highway agencies are currently using automated
methods for collecting pavement distress data. However, the process may vary among state
agencies, the basic understandings are quite similar in how the data are collected. Two of the
most prominent pavement distress parameters used by ArDOT are international roughness index
(IRI) and rutting to characterize pavement distress. In fact IRI and rutting are the only two
parameters ArDOT uses for rating pavements. The threshold values used by ArDOT for rating
pavements based on IRI and rutting is shown in Table 4.11.
TRC 1501 Final Report
Page | 55
Table 4.11: Threshold Values of Pavement Distresses for ArDOT
IRI – Asphalt
Scoring Rating
000 – 060 Very Good
060 – 095 Good
095 – 170 Fair
170 - Poor
Rutting
0.000 – 0.125 Excellent
0.125 – 0.350 Good
0.350 – 0.500 Fair
0.500 - Poor
The pavement management section of ArDOT is responsible for collecting, processing,
analyzing and reporting pavement performance data. ArDOT uses an Automated Road Analyzer
(ARAN) to collect IRI and rutting data. Figure 4.27 shows an ARAN used by ArDOT
Figure 4.27: Automated Road Analyzer
The ARAN is used to monitor pavement performance data and pavement imageries on
approximately 9500 centerline miles of roadways per year. Upon collection of these data, they
are delivered to the ArDOT computer servers where they are referenced through the ArDOT
geographic information system and processed with different analysis software (ArDOT, 2014).
The pavement management division of ArDOT uses a multimedia highway information system
(MMHIS) to report the data. Two hard drives containing the ARAN data were collected from
ArDOT to collect pavement distresses. These data were analyzed by MMHIS software. Figure
4.28 shows a screenshot of MMHIS software.
TRC 1501 Final Report
Page | 56
Figure 4.28: MMHIS Screenshot.
To check for any negative effects of PPA-modified asphalt binders, field data were
collected from Arkansas Interstate system to verify if they contained PPA. The brief
methodology of collecting field data is explained below:
Acid detection test was performed on the recovered binders from all ten sections of an
ArDOT sponsored Technical Research Committee (TRC) project (TRC 1404). The sections from
which the PPA-modified asphalt binder was detected were identified. Afterwards, the IRI and
rutting data for that particular section was collected by using MMHIS software.
TRC 1501 Final Report
Page | 57
5 Data and Analysis
5.1 Binder Performance Tests
5.1.1 Penetration Test Results
Penetration test results, presented in Figure 5.1, reveal that Source 2 binders are harder than
Source 1 binders at a room temperature of 25oC. With the addition of PPA, the penetration
values decrease. A similar trend is observed for the SBS modification. However, the decrease in
penetration values for PPA-modified binder is proportional to the amount of PPA being used in
the modification process. Furthermore, the PPA-modified PG 70-22 binder (S1B3 or S2B3) is
softer than its SBS-modified counterpart (S1B7 or S2B7). The addition of the LAA, Kao
Gripper™, softened the binders (i.e., increases penetration values), irrespective of source, as
indicated in penetration values of S1B4 and S2B4, as presented in Figure 5.1. Finally, the
increase in softness of the binder due to the addition of Kao Gripper™ is not prominent in the
case of Source 2 binder.
As mentioned earlier, at a room temperature of 25oC, the penetration values increased
slightly in the case of Kao Gripper®. While comparing penetration values of other four LAAs-
modified binders, an opposite trend was observed (Figure 5.2). The penetration values decreased
due to the addition of AD-here®
HP Plus, PermaTac Plus®, Evotherm
®, or PaveGrip
®. It is worth
recalling that the base binder used for LAA modification was PPA-modified PG 70-22 binders
(S1B3 and S2B3). Among these LAAs, Evotherm® showed the lowest penetration values, which
was followed by PermaTac Plus®, PaveGrip
®, and AD-here
® HP Plus. Overall, LAAs did not
have any adverse effects on the penetration grade of the binder.
TRC 1501 Final Report
Page | 58
Figure 5.1: Penetration Test Results.
Figure 5.2: Penetration Test Results for PPA+LAA Modified Samples.
5.1.2 Rotational Viscosity (RV) Test Results
RV data, shown in Table 5.1, followed the same pattern as the penetration values for binders
from both sources. As expected, PG 64-22 binder (e.g., B1S1 from Source 1) from any particular
source showed the lowest viscosity. As mentioned earlier, penetration data showed that binders
from Source 1 were softer than Source 2 at a room temperature. But, it was not the case for
viscosity values at higher temperatures. Binders from Source 2 showed considerably lower
viscosity values (softer) than the corresponding binders from Source 1. The viscosity of the
binders decreased upon the addition of LAA (Kao Gripper®).
From RV test data, the mixing and compaction temperatures for all 16 asphalt binder samples
were estimated as recommended by the Asphalt Institute (AI). As per AI, these temperatures
25
30
35
40
45
50
55
60
B1 B2 B3 B4 B5 B6 B7 B8
Pen
etra
tion
Nu
mb
er
Asphalt Binder Sample
S1 S2
25
30
35
40
45
50
55
60
Pavegrip PermaTac Adhere Evothrm Gripper B3
Pen
etra
tion
Nu
mb
er
LAA
S1 S2
TRC 1501 Final Report
Page | 59
should be determined where the viscosity‐temperature line crosses the viscosity ranges of 170
20 mPa.s (mixing temperature range) and 280 30 mPa.s (compaction temperature range). The
viscosity temperature line was determined using the procedure described in ASTM D2493,
“Standard Viscosity‐Temperature Charts for Asphalts.” Table 5.2 shows the mixing and
compaction temperatures of PG 64-22 (S1B1 and S2B1), PPA-modified PG 70-22 (S1B3 and
S2B3), SBS-modified PG 70-22 (S1B7 and S2B7) and PPA+SBS modified PG 76-22 (S1B8 and
S2B8) binders. As seen from Table 5.2, the mixing and compaction temperatures of SBS-
modified PG 70-22 binder (S1B7 or S2B7) are significantly higher than those of PPA-modified
PG 70-22 binder (S1B3 or B2B3). Thus, from the energy consumption perspective, PPA-
modified binders exhibit more favorable results than the SBS counterparts.
Table 5.1: Rotational Viscosity (mPa.s) Data of S1 and S2 samples
Binder
Type
Viscosity at Testing Temperature
135°C 150°C 165°C 180°C
S1B1 504.17 254.17 145.83 75.00
S1B2 595.83 287.50 150.00 75.00
S1B3 704.17 345.83 183.33 100.00
S1B4 554.00 270.83 145.83 75.00
S1B5 733.33 312.5 154.17 75.00
S1B6 733.33 325.00 162.50 75.00
S1B7 1271.00 595.67 312.50 175.00
S1B8 1929.33 870.67 450.00 262.50
S2B1 445.83 208.33 112.50 62.50
S2B2 570.83 270.83 133.33 75.00
S2B3 645.83 295.83 145.83 75.00
S2B4 620.83 304.17 145.83 75.00
S2B5 745.33 340.83 162.50 75.00
S2B6 704.17 341.67 187.50 100.00
S2B7 1271.00 554.17 279.17 162.5
S2B8 1767.00 758.33 350.00 187.50
TRC 1501 Final Report
Page | 60
Table 5.2: Mixing and Compaction Temperatures of PPA and SBS Modified Binders
Source 1
Binder Type
Mixing Temperature (°C) Compaction Temperature (°C)
High Low High Low
S1B1 165 158 150 145
S1B3 170 164 157 152
S1B7 183 177 171 165
S1B8 191 186 180 175
Source 2
Binder Type
Mixing Temperature (°C) Compaction Temperature (°C)
High Low High Low
S2B1 158 152 146 142
S2B3 164 159 154 149
S2B7 182 176 168 162
S2B8 185 180 173 168
5.1.3 Dynamic Shear Rheometer (DSR) Test Results
DSR tests were done in three aging conditions, unaged, RTFO-aged and PAV-aged in order to
characterize the viscoelastic behavior of asphalt binders at high and intermediate service
temperatures. Figures 5.3 through 5.6 show DSR test results of unaged and RTFO-aged asphalt
binders. As seen in these figures, all tested binders met the corresponding Superpave rutting
factor (G*/sin) criteria at their high PG temperatures (G*/sin should be at least 1.00 kPa for
unaged binders and 2.20 kPa for RTFO-aged binders). The dark horizontal lines in these figures
represent the Superpave acceptance criteria. It is clear that PPA-modified unaged and RTFO-
aged binders showed increased rutting factor (G*/sin) compared to the unmodified binders.
Moreover, SBS-modified binders indicated higher rutting resistance than the corresponding
PPA-modified PG 70-22 binders.
TRC 1501 Final Report
Page | 61
Figure 5.3: DSR Test Results of Unaged Binders from Source 1.
Figure 5.4: DSR Test Results of Unaged Binders from Source 2.
0.6
1.1
1.6
2.1
63 68 73 78 83 88
G*/s
inδ
(k
Pa)
Temperature °C
S1B1 Superpave Specification
S1B3 S1B4
S1B5 S1B2
S1B6 S1B7
S1B8
0.6
1.1
1.6
2.1
2.6
60 65 70 75 80 85
G*/s
inδ
(k
Pa)
Temperature °C
Superpave Specification S2B5 S2B1 S2B2 S2B3 S2B4 S2B6 S2B7 S2B8
TRC 1501 Final Report
Page | 62
Figure 5.5: DSR Test Results of RTFO-aged Binders from Source 1.
Figure 5.6: DSR Test Results of RTFO-aged Binders from Source 2.
As seen in Figures 5.7 through 5.10, the addition of LAA decreased the rutting resistance
of PPA-modified binders in both aging conditions. Among the LAAs, Gripper (S1B4-Gripper)
1.5
2.5
3.5
4.5
5.5
6.5
7.5
63 68 73 78 83
G*/s
inδ
(k
Pa)
Temperature °C
S1B1 Superpave Specification S1B3 S1B4 S1B5 S1B2 S1B7 S1B6
1.3
1.8
2.3
2.8
3.3
3.8
4.3
4.8
63 68 73 78 83
G*/s
inδ
(k
Pa)
Temperature °C
Superpave Specification S2B1
S2B2 S2B3
S2B4 S2B5
S2B6 S2B7
S2B8
TRC 1501 Final Report
Page | 63
was found to be the most compatible, whereas Adhere HP Plus was the least compatible. In the
case of the unaged condition and for Source 1, Gripper plotted slightly below the minimum value
of Superpave acceptance (Figure 5.7). But, for Source 2, the unaged Gripper-modified binder
passed the Superpave specification limit for G*/sinwhich should be at least 1.00 kPa. Under
the RTFO-aging condition, the Gripper-modifies binders from both sources passed the
Superpave specification limit for G*/sinwhich should be at least 2.20 kPa. The other LAAs-
modified binders from either source failed to meet the Superpave rutting criteria for both unaged
and RTFO-aging conditions. Thus, it is recommended to use a compatible LAA such as Gripper
when PPA is used as a modifier. This becomes an issue when contractors are given an option to
use LAA in asphalt mixes. However, it may not be a problem when refineries blend a compatible
LAA. Therefore, it is important for the ArDOT to know which LAA, if any, is used in the PPA-
modified binder.
Figure 5.7: DSR Test Results of Unaged PPA+LAA Modified Binders from Source 1.
0.5
0.75
1
1.25
1.5
1.75
2
61 64 67 70 73 76
G*/s
inδ
(k
Pa)
Temperature °C
S1B4-Pavegrip S1B4-PermaTac
S1B4-Adhere S1B4-Evotherm
S1B3 S1B4-Gripper
Superpave Specification
TRC 1501 Final Report
Page | 64
Figure 5.8: DSR Test Results for Unaged PPA+LAA Binders from Source 2.
Figure 5.9: DSR Test Results of RTFO-Aged PPA+LAA Binders from Source 1.
0.5
0.75
1
1.25
1.5
1.75
2
61 64 67 70 73 76
G*/s
inδ
(k
Pa)
Temperature °C
S2B4-Pavegrip S2B4-PermaTac S2B4-Evotherm S2B3 S2B4-Gripper Superpave Specification S2B4-Adhere
0
1
2
3
4
5
64 67 70 73 76
G*/s
inδ
(k
Pa)
Temperature °C
S1B4-Pavegrip S1B4-PermaTac S1B4-Adhere S1B4-Evotherm Superpave Specification S1B3 S1B4-Gripper
TRC 1501 Final Report
Page | 65
Figure 5.10: DSR Test Results of RTFO-Aged PPA+LAA Binders from Source 2.
DSR test results (Figures 5.11 and 5.12) on PAV-aged binders show fatigue
characteristics of tested asphalt binders. As required by Superpave specifications, the G*sin
value of a PAV-aged binder at the intermediate temperature should not be more than 5000 kPa.
The horizontal lines in these figures represent the Superpave maximum limit for fatigue
resistance of binders. Test results reveal that all tested binder samples met the Superpave fatigue
criterion. Test results also indicate that PPA-modified binders (S1B3 or S2B3) are more fatigue
resistant than the corresponding SBS-modified binders (S1B7 or S2B7). No specific trend
(increasing or decreasing) was observed in the case of foamed asphalt binders. In the case of
Source 1, the fatigue resistance of the foamed binder (S1B5) slightly decreased, whereas it
improved in the case of Source 2 binder. However, foamed binders (S1B5 and S2B5) from both
sources met the Superpave specified fatigue criterion.
0 0.5
1 1.5
2 2.5
3 3.5
4 4.5
64 67 70 73 76 79
G*/s
inδ
(k
Pa)
Temperature °C
S2B4-Pavegrip S2B4-PermaTac
S2B4-Adhere S2B4-Evotherm
S2B3 S2B4-Gripper
Superpave Specification
TRC 1501 Final Report
Page | 66
Figure 5.11: DSR Test Results of PAV-aged Binders from Source 1.
Figure 5.12: DSR Test Results of PAV-aged Binders from Source 2.
As seen from Figures 5.13 and 5.14, the addition of LAA in PPA-modified PG 70-22
binders did not cause them to fail the Superpave specification limit for fatigue even though
G*sin values increased in several cases. As a matter of fact, the fatigue resistance of the PPA-
modified binder from Source 1 increased when Gripper was used as the LAA. These findings
also indicate the need for a proper selection of a LAA for PPA-modified asphalts.
0 1000 2000 3000 4000 5000 6000 7000 8000
19 22 25 28
G*si
nδ
(k
Pa)
Temperature °C
S1B1 S1B2 S1B3 S1B4 S1B5 S1B6 S1B7 S1B8 Superpave Specification
0 1000 2000 3000 4000 5000 6000 7000 8000
19 22 25 28
G*si
nδ
(k
Pa)
Temperature °C
S2B1 S2B2
S2B3 S2B4
S2B5 S2B6
S2B7 S2B8
Superpave Specification
TRC 1501 Final Report
Page | 67
Figure 5.13: DSR Test Results of PAV-aged PPA+LAA Binders from S1.
Figure 5.14: DSR Test Results of PAV-aged PPA+LAA Binders from S2.
5.1.4 Bending Beam Rheometer (BBR) Test Results
BBR tests were performed to measure low temperature stiffness and stress relaxation properties
of asphalt binders. From the BBR test results, creep stiffness (S-value) and the slope of the
stiffness curve (m-value) at 60s were measured. It could be noted that all BBR tests were
conducted at 10oC higher than the low PG temperatures of the binders, as recommended by the
Superpave test specifications. For instance, in the case of PG 70-22 binders, BBR tests were
0
2000
4000
6000
8000
19 22 25 28
G*si
nδ
(k
Pa)
Temperature °C
S1B4-Pavegrip S1B4-PermaTac
S1B4-Adhere S1B4-Evotherm
S1B3 S1B4-Gripper
Superpave Specification
0
2000
4000
6000
19 22 25 28
G*si
nδ
(k
Pa)
Temperature °C
S2B4-Pavegrip S2B4-PermaTac
S2B4-Adhere S2B4-Evotherm
S2B3 S2B4-Gripper
Superpave Specification
TRC 1501 Final Report
Page | 68
conducted at -12oC. Superpave specifications require binder’s S-value should be not more than
300 MPa, and m-value should be at least 0.300.
S-values of tested binder samples are plotted in Figures 5.15 and 5.16. From these
figures, it is seen that all tested binders comfortably met the Superpave S-value criterion. At a
test temperature of -12°C, the lowest S-value for all tested binders from Source 1 is observed for
S1B8, which is a PPA+SBS modified PG 76-22 binder. Between PPA- and SBS-modified PG
70-22 binders from Source 1, S1B7 (SBS-modified) showed lower creep stiffness at -12°C than
the other. In the case of all binders from Source 2, the lowest creep stiffness was observed for
S2B4 (LAA+PPA-modified PG 70-22 binder).
Figure 5.15: Creep Stiffness of Binders from Source 1.
Figure 5.16: Creep Stiffness of Binders from Source 2.
40
60
80
100
120
-15 -12 -9 -6
Sti
ffn
ess
at
60 s
, M
Pa
BBR Test Temperature °C
S1B1 S1B2 S1B3 S1B4
S1B5 S1B6 S1B7 S1B8
40
60
80
100
120
-15 -12 -9 -6
Sti
ffn
ess
at
60 s
, M
Pa
BBR Test Temperature °C
S2B1 S2B2 S2B3 S2B4
S2B5 S2B6 S2B7 S2B8
TRC 1501 Final Report
Page | 69
As seen from Figures 5.17 and 5.18, all tested binder samples comfortably met the
Superpave m-value criterion at their low PG temperature (-22oC). At a testing temperature of -
12°C, the highest m-value was achieved for both S1B7 and S1B8. Their m-values at any
particular test temperature (-9°C or -12°C) are the same, and they overlap each other in the chart.
At -12°C, among all binders from Source 2, the highest m-value was found to be 0.4, which was
observed for S2B5 (PPA-modified foamed PG 70-22) and S2B6 (PG 64-22+1% PPA). For
Source 2, at -12°C, the same m-value was observed S2B3 and S2B7.
Figure 5.17: “m-values” of Binders from Source 1.
Figure 5.18: “m-values” of Binders from Source 2.
0.33
0.35
0.37
0.39
0.41
0.43
-15 -12 -9 -6
m-v
alu
e at
60 s
of
Load
BBR Test Temperature °C
S1B1 S1B2 S1B3 S1B4 S1B5 S1B6 S1B7 S1B8
0.33
0.35
0.37
0.39
0.41
0.43
-15 -12 -9 -6
m-v
alu
e at
60 s
of
Load
BBR Temperature °C
S2B1 S2B2 S2B3 S2B4
S2B5 S2B6 S2B7 S2B8
TRC 1501 Final Report
Page | 70
5.1.5 Moisture Susceptibility Analysis Using Surface Free Energy (SFE) Analysis
Moisture damage or stripping in asphalt mixtures is related to the adhesive bond energy and the
magnitude of reduction in this energy when the asphalt binder debonds from the aggregate
surface in the presence of water. It is believed that the affinity of the aggregates for water is
greater than that of the asphalt binder. During the stripping process, water diffuses through the
thin film of asphalt binder or mastic and collects on the surface of the aggregate. Therefore, the
estimation of these adhesive energies has the potential to be used as a screening tool for material
selection. Further, it is important to understand the effect of PPA modifications on moisture
susceptibility of binder aggregate systems through a fundamental science approach, called the
surface science.
Moisture susceptibility of asphalt binders was measured by conducting the SFE analysis.
The SFE could be defined as the amount of energy produced by a new interface inside a vacuum.
SFE tests were conducted on unaged binders at a room temperature. The first part of the SFE
analysis was to measure the static contact angles of the asphalt binders and aggregate samples
using three probe liquids (water, ethylene glycol, and formamide) of known SFE components.
Four aggregates were chosen for the SFE analysis in this study. Two of these aggregates were
sandstone (Preston Sandstone) and gravel (Preston Gravel) from Arkansas, and SFE values of
two other aggregates (Snyder Granite and Martin Marietta Mill Creek [MMMC] Granite) were
chosen from literature for comparison.
From the contact angles of asphalt binders and aggregate samples, the SFE components,
namely, a mono-polar acidic component (Γ+), a mono-polar basic component (Γ
-), and an apolar
or Lifshitz-van der Waals component (ΓLW
) were calculated. Figures 5.19 and 5.20 show the
contact angles of the asphalt binder samples coated on thin glass slides. In general, among three
probe liquids, water made the highest contact angles with the asphalt binder samples. Between
binders from Source 1 and Source 2, the samples from the former had higher contact angles than
those from the latter, which was expected as binders from Source 2 binders were stiffer than
those of Source 1 at a room temperature.
TRC 1501 Final Report
Page | 71
Figure 5.19: Contact Angles of Binders from Source 1.
Figure 5.20: Contact Angles of Binders from Source 2.
Table 5.3 shows the SFE parameters along with work of cohesion (WOC) values for
asphalt binders. In this table, Γab
is the acid-base component of the SFE, and it is the geometric
average of Γ+ and Γ
- . It could be noted that the WOC is applicable for asphalt binders, and it is
twice of Γtotal
, which is the sum of Γab
and ΓLW
. The higher the WOC of an asphalt binder, the
more energy is needed to break its cohesive bonds. The SFE values of aggregates are irrelevant
while estimating asphalt binder’s WOC. For Source 1 binders, the highest WOC was observed
from S1B8, which was a combination of PPA- and SBS-modified binder. But, none of the
binders modified solely by PPA had higher cohesion energy compared to their base binders (PG
64-22 aka S1B1). Rather, for Source 1, all binders modified by PPA alone showed less cohesion
energy than S1B1. While comparing PPA-modified (S1B3) and SBS-modified (S1B7) binders,
S1B7 had higher cohesion energy than S1B3. In Source 2, S2B3 possessed an increased WOC
compared to S2B1 and S2B7. However, the WOC of S2B6 (1.0% PPA) decreased compared to
60
70
80
90
100
Water Ethylene Glycol Formamide
Con
tact
An
gle
(D
egre
e)
Probe Liquid
S1B1 S1B2 S1B3 S1B4 S1B5 S1B6 S1B7 S1B8
60
70
80
90
100
110
Water Ethylene Glycol Formamide Con
tact
An
gle
(D
egre
e)
Probe Liquid
S2B1 S2B2 S2B3 S2B4 S2B5 S2B6 S2B7 S2B8
TRC 1501 Final Report
Page | 72
S2B3 (0.75% PPA), indicating a decreasing trend of bond energy beyond the optimum dosage of
PPA. Another interesting observation was that the addition of LAA decreased the cohesion
energy of S2B3, defeating the purpose of adding LAA. It could be noted that adhesion bonds
between the aggregates and binder of an asphalt mixture are dominating parameters over the
cohesive bonds for achieving favorable moisture resistance, and a LAA is added to improve the
adhesive bonds. The adhesive bonds between aggregates and asphalt binders are discussed in the
subsequent section.
Table 5.3: SFE Parameters (mJ/m2) and Cohesion Energy (mJ/m
2) of asphalt binders
Probe Liquid/
Test Sample
Γ+ Γ
- Γ
LW Γ
ab Γ
total W
CL
Water 25.50 25.50 21.80 - 72.80 N/A
Ethylene Glycol 1.92 47.00 29.00 - 48.00 N/A
Formamide 2.28 39.60 39.00 - 58.00 N/A
Snyder Granite 0.10 8.43 35.15 1.87 37.03 N/A
MMMC Granite 0.42 36.98 35.84 7.89 43.73 N/A
Preston Gravel 20.93 14.95 13.75 35.37 49.12 N/A
Preston Sandstone 20.76 14.76 13.56 35.00 48.56 N/A
S1B1 12.61 2.56 0.94 11.36 12.30 24.60
S1B2 12.50 2.30 0.18 10.72 10.90 21.80
S1B3 12.50 2.31 0.43 10.74 11.17 22.34
S1B4 12.52 2.35 0.53 10.84 11.37 22.74
S1B5 12.54 2.39 0.62 10.94 11.56 23.12
S1B6 12.51 2.33 0.50 10.79 11.29 22.58
S1B7 12.92 3.34 2.90 13.13 16.03 32.06
S1B8 13.23 4.00 3.55 14.54 18.09 36.18
S2B1 12.70 2.63 2.15 11.55 13.70 27.40
S2B2 12.59 2.36 1.87 10.90 12.77 25.54
S2B3 13.02 3.44 3.01 13.38 16.39 32.78
S2B4 12.72 2.81 2.38 11.95 14.33 28.66
S2B5 12.61 2.57 2.14 11.38 13.52 27.04
S2B6 12.37 2.00 1.54 9.94 11.48 22.96
S2B7 12.67 2.83 2.44 11.97 14.41 28.82
S2B8 12.67 2.83 2.44 11.97 14.41 28.82
TRC 1501 Final Report
Page | 73
The variation of adhesion energy under the dry condition (Gdry) is shown in Table 5.4.
The work of adhesion is defined as the amount of energy necessary to separate two materials at
their interface. As seen from this table, MMMC granite showed higher Gdry values than the
others, irrespective of the binder types. Among all tested aggregates, under the dry condition,
SBS- and PPA-modified PG 76-22 binder (S1B8) from Source 1 showed the highest Gdry value.
In the case of Source 2, PPA-modified PG 70-22 binder (S2B3) showed the highest Gdry. The
Gdry values between Preston gravel and Preston sandstone did not have significant differences,
but they were notably lower than those of MMMC granite. The SFE data also shows that an
increase in the PPA amount did not increase the Gdry value for Preston’s gravel or sandstone. It
can be noted that the higher the Gdry, the better adhesive bonds exist between aggregates and
binder under the dry condition. It is the opposite in the case of the adhesion energy under the wet
condition (ΔGwet) as values presented in Table 5.5 are negative.
Table 5.4: Work of Adhesion (mJ/m2) for Asphalt-Aggregate System in Dry Condition
Binder Sample
Aggregates
Preston Gravel Preston
Sandstone Snyder Granite
MMMC
Granite
S1B1 34.7 34.4 32.1 54.8
S1B2 30.5 30.3 25.6 48.1
S1B3 32.2 32.0 28.3 50.9
S1B4 32.8 32.5 29.2 51.8
S1B5 33.2 33.0 29.9 51.8
S1B6 32.6 32.4 28.9 51.5
S1B7 40.4 40.2 41.1 64.1
S1B8 42.1 41.8 43.5 66.8
S2B1 38.4 38.2 38.1 60.9
S2B2 37.6 37.3 36.8 59.5
S2B3 40.8 40.5 41.5 64.7
S2B4 39.0 38.8 39.0 61.8
S2B5 38.3 38.1 38.0 60.7
S2B6 36.4 36.2 35.1 57.6
S2B7 39.1 38.9 39.2 62.0
S2B8 39.1 38.9 39.2 62.0
The ΔGwet is a measure of adhesion energy under the wet condition and the negative
values suggest the de-bonding potential of asphalt binders and aggregates in the presence of
TRC 1501 Final Report
Page | 74
water. From Table 5.5, it could be seen that the addition of PPA slightly decreased (as values are
negative) the ΔGwet of the binders from Source 1, which would make the binder more moisture
susceptible. On the other hand, binders from Source 2 acted differently with the addition of PPA.
For instance, the addition of 0.75% PPA (S2B3) slightly increased the ΔGwet of the base binder
(S2B1) from -12.9 mJ/m2 to -12.1 mJ/m
2, indicating an increased resistance to stripping.
Furthermore, LAA was found to be ineffective in reducing the negative bond energy under the
wet condition. It could be noted that the combination of adhesion energy values under both dry
and wet conditions rather than only dry or wet condition would have to be considered in
determining the compatibility between aggregates and binders and a get a better understanding of
their stripping resistance. The term “compatibility ratio,” introduced by Texas Transportation
Institute (TTI), is illustrated next.
Table 5.5: Work of Adhesion (mJ/m2) for Asphalt-Aggregate System in Wet Condition
Binder Sample
Aggregates
Preston Gravel Preston
Sandstone Snyder Granite
MMNC
Granite
S1B1 13.9 14.3 29.8 17.5
S1B2 15.1 15.5 29.3 16.8
S1B3 14.6 15.1 29.8 17.4
S1B4 14.5 14.9 29.9 17.5
S1B5 14.4 14.8 29.9 17.5
S1B6 14.5 14.9 29.9 17.5
S1B7 12.2 12.5 29.3 17.5
S1B8 11.6 11.9 27.9 16.6
S2B1 12.9 13.3 30.8 18.7
S2B2 13.2 13.6 31.4 19.1
S2B3 12.1 12.4 29.0 17.4
S2B4 12.7 13.1 30.4 18.4
S2B5 13.0 13.3 31.0 18.8
S2B6 13.7 14.1 32.4 19.8
S2B7 12.7 13.1 30.4 18.4
S2B8 12.7 13.1 30.4 18.4
The “compatibility ratio” of an asphalt binder and aggregate system indicates the
potentiality of moisture resistance of the binder with the aggregate. A higher compatibility ratio
(CR) means the binder and aggregate system is less vulnerable to moisture damage. It is the ratio
TRC 1501 Final Report
Page | 75
of ΔGdry and -ΔGwet. Generally, the CR increases if the ΔGdry increases and/or ΔGwet decreases,
and vice versa. In particular, a CR value of less than 0.5 is considered to be very poor, whereas
CR values of more than 0.5 signify good compatibility between binder and aggregates. To be
more precise if the CR value is greater than 1.5 the compatibility is rated “very good” and it is
graded as an “A.” The range of CR between 0.5 and 1.5 means “good” and graded as “B”
whereas and CR values between 0.5 and 0.75 means “poor” and graded as “C.” Furthermore, CR
values less than 0.5 means “very poor” compatibility and graded as “D.” As suggested by the
TTI researchers, the qualitative description of compatibility is shown in Table 5.6.
Table 5.6: Qualitative Description of Compatibility
Compatibility Ratio Range Grading
Greater Than 1.5 A (Good)
0.75 – 1.5 B
0.5 – 0.75 C
Less Than 0.5 D (Poor)
The compatibility analysis of tested asphalt binder samples from Source 1 and Source 2
are shown in Figures 5.21 and 5.22, respectively. From these figures, it is seen that the CR values
of the binder aggregate systems ranged from “B” to “A.” Further, it is see that the two aggregates
(Preston’s gravel and sandstone) collected from Arkhola, AR showed very close CR. With the
addition of 0.5% PPA, which was the optimum amount of PPA for Source 1, yielded
significantly higher CR values than other aggregates. From the CR values of binders from
Source 2, it was clear that both gravel and sandstone had very similar CR values. On the other
hand, MMMC granite showed the highest CR values and Snyder granite showed the lowest.
Moreover, for Source 2, PPA-modified PG 70-22 (S2B3) showed higher CR values than either
PPA+SBS modified PG 76-22 (S2B8) or SBS-modified PG 70-22 (S2B7).
TRC 1501 Final Report
Page | 76
Figure 5.21: Compatibility Ratio of Binders from Source 1.
Figure 5.22: Compatibility Ratio of Binders from Source 2.
The variations of CR with varying PPA contents are shown in Figures 5.23 and 5.24. As
seen from these figures, there is no significant change in CR values with the addition of PPA in
Source 1 binders. However, for Source 2 binders, it is clear that the CR value is the highest at the
optimum amount of PPA, but gradually decrease afterward. This indicates a reduction of the
effectiveness of PPA in term of stripping resistance when higher than the optimum amount is
used during the modification process.
0 0.5
1 1.5
2 2.5
3 3.5
4 4.5
5
S1B1 S1B2 S1B3 S1B4 S1B5 S1B6 S1B7 S1B8
Com
pati
bil
ity R
ati
o
Asphalt Binder Sample
Preston Gravel Preston Sandstone
Snyder granite Martin-Marietta-Mill-Creek granite
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
S2B1 S2B2 S2B3 S2B4 S2B5 S2B6 S2B7 S2B8
Com
pati
bil
ity R
ati
o
Asphalt Binder Sample
Preston Gravel Preston Sandstone
Snyder granite Martin-Marietta-Mill-Creek granite
TRC 1501 Final Report
Page | 77
Figure 5.23: Variation of CR Values with Different Dosage of PPA for Source 1.
Figure 5.24: Variation of CR Value with Different Dosage of PPA for Source 2.
The CR values of PPA+LAA modified binders are shown in Table 5.7. From CR data
presented in this table, it is evident that the CR values are in the compatibility category of “B” or
“A.” However, for Source 1, the CR value reduced significantly when AD-Here HP Plus was
used as the LAA. This alluded to a hypothesis that if a “C” category of binder aggregate system
was modified with AD-Here HP Plus, its CR would degrade to “D.” However, MMMC granite
was found have the same or increased CR values with all LAAs. In the case of Source 2 and any
of the Preston aggregates (gravel and sandstone), the CR value decreases significantly for all
LAAs except Kao Gripper®
X2. However, MMMC granite was found have the increased CR
0
2
4
-0.25 3E-15 0.25 0.5 0.75
Com
pati
bil
ity R
ati
o
(CR
)
% PPA (For Source 1)
Preston Gravel Preston Sandstone Snyder granite Martin-Marietta-Mill-Creek granite
0.0
1.0
2.0
3.0
4.0
0 0.25 0.5 0.75 1 Com
pati
bil
ity R
ati
o (
CR
)
% PPA (For Source 2)
Preston Gravel Preston Sandstone
Snyder granite Martin-Marietta-Mill-Creek granite
TRC 1501 Final Report
Page | 78
values with PermaTac Plus and AD-Here HP Plus. Such observations reiterate the need for a
careful selection of LAA when PPA is used as a modifier.
Table 5.7: Compatibility Ratio of PPA+LAA Modified Binders
Source 1
Preston
Gravel
Preston
Sandstone
Snyder
Granite MMMC Granite
S1B4-Pavegrip 2.09 2.02 1.07 3.97
S1B4-Permatac 1.70 1.64 0.87 3.18
S1B4-Adhere 2.19 2.12 1.12 4.15
S1B4-Evotherm 2.18 2.11 1.11 4.10
S1B3 2.21 2.12 0.95 2.93
S1B4 (Gripper) 2.26 2.18 0.98 2.93
Source 2
Preston
Gravel
Preston
Sandstone
Snyder
Granite MMMC Granite
S2B4-Pavegrip 1.79 1.73 0.92 3.33
S2B4-Permatac 2.23 2.15 1.14 4.20
S2B4-Adhere 2.36 2.28 1.21 4.43
S2B4-Evotherm 1.67 1.61 0.85 3.05
S2B3 3.37 3.27 1.43 3.72
S2B4 (Gripper) 3.07 2.96 1.28 3.36
5.1.6 Elastic Recovery (ER) and Multiple Stress Creep Recovery (MSCR)
Elastic recovery (ER) tests of the unaged binder samples were performed at ArDOT Materials
Division under the guidance of Ms. Dawn Richards. The ER tests were done in accordance with
AASHTO T 301. As expected, very high ER values were observed for binders containing SBS or
in a combination of SBS and PPA. The corresponding ER values for S1B7, S1B8, S2B7, and
S2B8 were found to be 85%, 84%, 88.5%, and 84%, respectively. The base binder samples
(S1B1 and S2B1) did not show any noticeable ER, which was also expected. However, very low
ER values were observed for binders containing PPA (no SBS) or PPA along with LAA. The
corresponding ER values of S1B2, S1B3, S1B4, S1B5, and S1B6 binder samples were found to
be 5.5%, 6.5%, 7.0%, 7.5%, and 8.5%, respectively. Similarly, the ER values of S2B2, S2B3,
S2B4, S2B5, and S2B6 binder samples were found to be 6.5%, 7.0%, 7.0%, 10.5%, and 10.0%.
TRC 1501 Final Report
Page | 79
The MSCR tests were conducted on RTFO-aged binder samples at A-State Asphalt Binders
Laboratory. The MSCR tests were conducted in accordance with AASHTO T 350. Similar to
ER, very low MSCR % Recovery values were observed for PPA-modified binders. The MSCR
percent recovery values of S1B1, S1B2, S1B3, and S1B6 were found to be 1.91%, 4.26%,
2.46%, and 9.91%. Based on Jnr values, the MSCR grades of S1B1, S1B2, S1B3, and S1B6 were
PG 64S-XX, PG 64H-XX, PG 64S-XX, and PG 64H-XX, respectively. Binders from Source 2
showed similar behavior. MSCR % Recovery values of S2B1, S2B2, S2B3, and S2B6 were
found to be 1.38%, 3.35%, 4.03%, and 8.91%, respectively. Furthermore, the MSCR grade of
any of these binders (S2B1, S2B2, S2B3, or S2B6) was estimated as PG 64H-XX. These test
results suggest that neither ER nor MSCR test is capable of characterizing PPA-modified
binders’ creep and recovery behavior. It could be noted that both ER and MSCR tests were
meant to be used for characterizing polymer-modified binders. The former is a PG Plus test, and
the latter is meant to be a replacement of the ER test method. Thus, it is recommended that an
alternate be explored for proper investigation of PPA-modified binders.
5.2 Chemical Test Results
5.2.1 Acidity Measurement Test
This test is useful in understanding the level of acidity of the binder, and thereby selecting
compatible aggregates for the asphalt mixture. Asphalt binders were tested for the acid number
to check the level of acidity after PPA- and/or SBS-modification. Figure 5.25 shows acid number
data for binders from both sources. The test results revealed that the neat binder from Source 1
was inherently basic and that from Source 2 acidic. The acidity increases (pH decreases) with the
addition of PPA, which is expected as PPA is an acid. It is also observed that the foaming (S1B3
or S2B3) of the PPA-modified binder (S1B5 or S2B5) reduces its acidity.
TRC 1501 Final Report
Page | 80
Figure 5.25: pH of Asphalt Binders.
5.2.2 Saturates Aromatics Resins and Asphaltenes (SARA) Analysis
The outcomes of the chromatographic separation of both sets of binders were analyzed. The
results have been presented in Figures 5.26 and 5.27. Asphalt binder samples from the Canadian
crude source, i.e. Source 1, had a high asphaltenes content (15%) compared to those (12.8%)
from the Arabian crude source, i.e. Source 2. The asphaltenes content increases and resins
content decreases with the addition of PPA, which makes the binder stiffer than its base binder.
These observations are in agreement with the findings of rheological data presented earlier in this
report.
0
1
2
3
4
5
6
7
8
9
S1B1 S1B2 S1B3 S1B4 S1B5 S1B6 S2B1 S2B2 S2B3 S2B4 S2B5 S2B6
pH
Asphalt Binder Samples
TRC 1501 Final Report
Page | 81
Figure 5.26: SARA Fractions of Source 1 Asphalt Binders.
Figure 5.27: SARA Fractions of Source 2 Asphalt Binders.
4.3 6.8 9.3 8.7 9.6 6.4 5.3 4.4
59.1
59.4
58.1
55.7
56.7
59.1
53.0
53.0
21.6
17.4
14.8
17.6
14.7
16.8
24.2
22.0
15.0
16.5
17.8
18.1
18.9
17.7
17.5
20.7
0%
20%
40%
60%
80%
100% S
1B
1
S1B
2
S1B
3
S1B
4
S1B
5
S1B
6
S1B
7
S1B
8
% F
ract
ion
s SARA Fractions of Asphalt Binders (Source 1)
Asphaltenes Resins Aromatics Saturates
4.8 4.7 6.3 6.8 7.4 6.6 6.5 8.3
68.1
67.3
63.4
62.5
63.3
63.2
62.7
62.3
14.3
13.0
14.6
13.7
12.1
12.6
15.7
13.1
12.8
15.0
15.7
17.0
17.2
17.6
15.1
16.2
0%
20%
40%
60%
80%
100%
S2B
1
S2B
2
S2B
3
S2B
4
S2B
5
S2B
6
S2B
7
S2B
8
% F
ract
ion
s
SARA Fractions of Asphalt Binders (Arabian Crude)
Asphaltenes Resins Aromatics Saturates
TRC 1501 Final Report
Page | 82
5.2.3 Fourier Transform Infrared Spectroscopy (FTIR)
FTIR analysis is a common and quick technique to identify the functional groups present in
asphalt binders. It is commonly used in the asphalt industry to identify the presence of any
specific functional group in asphalt binders. The main purpose of performing the FTIR tests (the
IR card method) was to observe any differences in the peaks due to the addition of PPA in
asphalt binders. Figure 5.28 shows the FTIR spectra of S1B1 and S1B2, and Figure 5.29 shows
the FTIR spectrum for PPA-modified binders from S1.
Figure 5.28: FTIR Spectrum of PG 64-22 Binders from S1 and S2.
At wavenumber between 800 and 810 cm-1
, a peak was located for all three binder
samples. This marks the presence of alkyl halides. It could be noted that the absorbance of this
peak for S1B2 increases with an increasing amount of PPA, but for more than the optimum
amount, which was 0.5% for this source, the absorbance reduced. This peak could be described
as the control functional group for unmodified binders. Moreover, the peak from 1740 -1600 cm-
1 signifies a strong presence of –OH. On the other hand, the peak found in the region of 1380-
1395 cm-1
strongly marks the presence of aromatics (= C-H). The second analyzed peak was
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
450 1450 2450 3450
Ab
sorb
an
ce
Wavenumber cm-1
S1B1 S2B1
TRC 1501 Final Report
Page | 83
found at 1595 to 1600 cm-1
, which marks the presence of aromatics as well, prompting to call
this as a predominant functional group for PPA-modified asphalt.
Figure 5.29: FTIR Spectra of PPA-modified Binders from Source 1.
Figure 5.30 shows the FTIR spectra of PPA-modified binders from Source 2. As seen in
this figure, a peak was found at 800 to 810 cm-1
in Source 2 binders as well. However, unlike
Source 1, S2B6 showed the highest absorbance in Source 2, whereas S2B2 showed the lowest.
Furthermore, a similar dip was also found at 1540 cm-1
in this source. Similar to the mechanistic
tests, FTIR tests were also performed on the LAA modified binders.
Figures 5.31 and 5.32 show the FTIR spectra of PPA+LAA modified binders from
Sources 1 and 2, respectively. The peaks for these binders are in agreement with the data
presented earlier. However, higher absorbance values are observed at 1595 to 1600 cm-1
cm-1
. At
a wavenumber of 1380-1395 cm-1
, a reasonably lower peak than the PPA-modified binder is
observed.
Figures 5.33 and 5.34 show the FTIR spectra for S2B7 and S2B8, respectively. FTIR
analysis of polymer modified samples shows peaks at 965cm-1
, attributed to SB and SBS. The
ratio of the SB and SBS peak versus the asphalt peak is then used to determine the polymer
content of the asphalt.
0
0.5
1
1.5
2
2.5
3
3.5
4
450 950 1450 1950 2450 2950 3450 3950
Ab
sorb
an
ce
Wavenumber cm-1
S1B3 S1B2 S1B6
TRC 1501 Final Report
Page | 84
Figure 5.30: FTIR Spectra of PPA-modified Binders from Source 2.
Figure 5.31: FTIR Spectrum for PPA+LAA binders from Source 1.
-0.5
0.5
1.5
2.5
3.5
4.5
5.5
6.5
450 1450 2450 3450
Ab
sorb
an
ce
Wavenumber cm-1
S2B3 S2B2 S2B6
0
1
2
3
450 950 1450 1950 2450 2950 3450 3950
Ab
sorb
an
ce
Wavenumber cm-1
S1B4-Adhere S1B4-Evotherm S1B4-Permatac S1B4-Pavegrip S1B4
TRC 1501 Final Report
Page | 85
Figure 5.32: FTIR Spectrum for PPA+LAA Modified Binders from Source 2.
Figure 5.33: Polymer Content Analysis of S2B7.
0
1
2
3
450 950 1450 1950 2450 2950 3450 3950
Ab
sorb
an
ce
Wavenumber cm-1
S2B4-Adhere S2B4-Evotherm S2B4-Permatac
S2B4-Pavegrip S2B4
0
1
2
3
4
5
6
600 800 1000 1200 1400 1600
Ab
sorb
an
ce
Wavenumber cm-1
TRC 1501 Final Report
Page | 86
Figure 5.34: Polymer Content Analysis of S2B8.
The absorbance and area analysis for S2B7 (2% SBS) and S2B8 (2% SBS and 0.75%
PPA) is shown in Table 5.8. As seen in this table the absorbance ratios of S1B7 and S1B8 are
0.35 and 0.10, respectively, which correspond to a corresponding polymer content of 5%, and
1.85%. However, the SBS content in these binders was 2%. According to the area analysis, the
area rations of S1B7 and S1B8 are 0.25 and 0.13, respectively, indicating the corresponding
polymer contents of 3.5% and 2%. Thus, the area analysis appears to be a better approach than
the absorbance analysis to predict the polymer content.
Table 5.8: Absorbance and Area Analysis of S2B7 and S2B8
Sample Absorbance Area Absorbance Area Absorbance
Ratio
Area
Ratio Peak 966 Peak 1375
S1B7 0.428 8.463 1.237 34.432 0.35 0.25
S1B8 0.382 6.552 3.526 51.334 0.10 0.13
Under the supervision of Ms. Dawn Richards, the ArDOT Materials Division personnel
also ran quantitative testing for the polymer content by FTIR on the samples that were indicated
as having 2% SBS. This testing was done in accordance with AASHTO T 302. The SBS
contents in S1B7, S1B8, S2B7, and S2B8 samples were found to be 1.97%, 2.07%, 1.96%, and
2.19%, respectively. Interestingly, after heating some of the cans and mixing them thoroughly,
0
1
2
3
4
5
6
7
600 800 1000 1200 1400 1600
Ab
sorb
an
ce
Wavenumber cm-1
TRC 1501 Final Report
Page | 87
consistent and expected results were observed. The repeatability of the test method was
supposed to be +/- 0.2%, which also found in this study.
5.3 Mixture Performance Tests
5.3.1 Evaluation of Rutting and Stripping of Asphalt (ERSA) Test
The ERSA test determines the susceptibility of premature failure of HMA caused by weakness in
the aggregate structure, inadequate binder stiffness, or moisture damage. The ERSA test is
essentially identical to the Hamburg Wheel Test. The wheel has a load of 158 lbs, and it should
make 52 passes across the specimen per minute with a maximum speed of 0.305 m/s. The
machine is limited to running 20,000 cycles. Arkansas specifications for surface courses require
a maximum rut depth of 8.000 mm at 8,000 cycles for an APA style wheel tracking tests. For
ERSA test, the maximum cycle value of 8,000 cycles and a maximum rut depth of 8.0 mm are
utilized.
The ERSA test was done for five mixtures, namely, S1B1, S1B3, S1B4, S1B5, and S1B7.
The aggregates were from Van Buren and are known to be moisture susceptible. The summary
of ERSA test results is shown in Figure 5.35. For S1B1, the numbers of passes required for 8.0
mm rut depth were 13,372 (6,686 cycles). After 16,000 passes (8,000 cycles), the rut depth was
10.26 mm. In the case of the foamed PPA-modified PG 70-22 mixture (S1B5), the passes at to
8.0 mm were 10,494 (5,247 cycles). The PPA-modified PG 70-22 mixture (S1B3) reached a rut
depth of 1.48 mm at 8000 cycles. The rut depths for S1B4 and S1B7 mixtures were 1.18 and 2.7
mm, respectively. On the other hand, for the PPA-modified PG 70-22 mixture (S1B3),
PPA+LAA modified mixture (S1B4), and SBS-modified PG 70-22 (S1B7), after 16,000 passes
(8,000 cycles), the rut depth of 8.0 mm was not reached (Table 5.9). In another way, the number
of passes to reach a rut depth of 8 mm for these mixes (S1B3, S1B4, and S1B7) were more than
16,000. Therefore, it is demonstrated that adding either PPA or PPA+LAA to the binder helps to
improve the rutting resistance of the mixture by more than a 100%.
TRC 1501 Final Report
Page | 88
Figure 5.35: Summary of ERSA Test Results.
Table 5.9: Evaluation of Rutting and Stripping of Asphalt Mixtures
Mixture with Binder
Sample
No. of Passes to 8.000 mm No. of Cycles to 8.000 mm
Left Right Average Left Right Average
PG 64-22 (S1B1) 11,682 15,060 13,371 5,841 7,530 6,686
PG 64-22 + 0.5% PPA
(S1B3) >16,000 >16,000 >16,000 >8,000 >8,000 >8,000
PG 64-22 + 0.5% PPA
+ 0.5% LAA (S1B4) >16,000 >16,000 >16,000 >8,000 >8,000 >8,000
PG 64-22 + 2% SBS
(S1B5) >16,000 >16,000 >16,000 >8,000 >8,000 >8,000
5.3.2 Uniaxial Dynamic Modulus
Dynamic modulus is often used to rank rutting and cracking characteristics of asphalt concrete.
In order to measure the Dynamic Modulus of an asphalt concrete sample, a dynamic load is
applied at different frequencies: 0.1 Hz, 0.5 Hz, 1.0 Hz, 5 Hz, 10 Hz, and 25 Hz. Samples are
tested at five different temperatures: -10 , 4 , 21 , 37 , and 54 . The amplitude of the
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0
2
4
6
8
10
S1B1 S1B3 S1B4 S1B7 S1B5
Cycl
es
Ru
t D
epth
, m
m
Binder Type
Rut Depth Cycles
TRC 1501 Final Report
Page | 89
load as well as the vertical deformation of the specimen using three extensometers is recorded.
The ratio of the amplitudes of vertical stress to vertical strains is then computed in order to
obtain the value of dynamic modulus for each combination of frequency and temperature by
following AASHTO T62-07. Specimens of 100 mm diameter and 150 mm height were used.
Figure 5.36 shows master curves for five mixtures, and Figure 5.37 shows the shift factors.
While the three master curves had a relatively similar dynamic modulus, PG 64-22 (S1B1), PPA-
modified PG 70-22 (S1B3), PPA+LAA modified (S1B4), PPA-modified foamed (S1B5) and
SBS-modified (S1B7) mixtures literally lie on top of each other, while the S1B5 shows a slightly
higher stiffness at intermediate reduced frequencies. Additionally, the SBS-modified PG 70-22
(S1B7) mixture shows higher stiffness in the low frequencies and lower stiffness in the higher
frequencies. Lower stiffness at low frequencies indicates a lower level of cracking susceptibility
while lower stiffness at high frequencies indicates a higher level of rutting susceptibility.
Figure 5.36: Uniaxial Dynamic Modulus Master Curves Summary.
0
3000
6000
9000
12000
15000
1.0E-05 1.0E-03 1.0E-01 1.0E+01 1.0E+03 1.0E+05 1.0E+07 1.0E+09
Dyn
am
ic M
od
ulu
s (|
E*|, M
Pa)
Reduced Frequency (Hz)
S1B1 S1B3 S1B4 S1B7 S1B5
TRC 1501 Final Report
Page | 90
Figure 5.37: Shift Factor – Uniaxial Dynamic Modulus.
5.3.3. Semi-Circular Bend (SCB) Fracture Test
Semi-Circular Bend (SCB) Test determines the fracture energy and fracture toughness at low
temperatures of asphalt mixtures by its semi-circular geometry. Figures 5.38 and 5.39 show the
fracture energy in J/m2 of each material at -10°C and -24°C, respectively. In Figure 5.38, S1B3
shows greater fracture energy than the other materials at -10°C, indicating the 0.5% PPA
increased cracking resistance at low temperatures greater than the low temperature binder grade.
However, in Figure 5.39, S1B7 shows greater fracture energy than the other materials at -24°C,
indicating the SBS modification increased cracking resistance at low temperatures lower than the
low temperature binder grade.
-4
-2
0
2
4
6
8
10
-20 0 20 40 60
Log S
hif
t F
act
or
Temperature (°C)
S1B1 S1B3 S1B4 S1B7 S1B5
TRC 1501 Final Report
Page | 91
Figure 5.38: Summary of SCB Test Fracture Energy at -10°C.
Figure 5.39: Summary of SCB Test Fracture Energy at -24°C.
5.3.4 IDT Creep Compliance and Indirect Tensile Strength
Creep compliance is defined as the time-dependent strain divided by the applied stress and
indirect tensile strength is the strength shown by a specimen that is subject to tension. The
purpose of the test was to determine the tensile creep by the application of a static load of fixed
magnitude along the diametrical axis of the specimen for the duration of 100 seconds. During the
creep test, loads were selected to remain horizontal strain in the linear viscoelastic range, which
0
100
200
300
400
500
600
700
800
S1B1 S1B3 S1B4 S1B7 S1B5
Fra
ctu
re E
ner
gy, J/m
2
Binder Type
0
200
400
600
800
1000
1200
S1B1 S1B3 S1B4 S1B7 S1B5
Fra
ctu
re E
ner
gy, J/m
2
Binder Type
TRC 1501 Final Report
Page | 92
was typically below of 500 x 10-6
mm/mm. When the creep tests were finished at each
temperature, the tensile strength was determined by applying a load to the specimen at a rate of
12.5mm of ram movement per minute. The IDT tensile strengths at -10°C for all four samples are
shown in Figure 5.40. The PPA-modified foamed PG 70-22 binder S1B5 showed the highest
tensile strength among the five tested mixtures, indicating the highest cracking resistance.
Furthermore, between the two PG 70-22 binders, the PPA-modified (S1B3) mixture showed
higher tensile strength than the SBS-modified (S1B7) mixture.
Figure 5.40: Summary of IDT Tensile Strength at -10°C.
5.3.5 Tensile Strength Ratio
The purpose of tensile strength ratio (TSR) (AASHTO T 283) test was to assess the effects of
saturation and accelerated water conditioning with a freeze-thaw cycle, of compacted asphalt
mixtures. The tensile strength ratios of five tested mixture samples are shown in Figure 5.41.
Between the two PG 70-22 mixtures, the SBS-modified (S1B7) mixture showed higher tensile
strength under both wet and dry conditions. Moreover, the tensile strength results of LAA-
modified mixtures did not show any improvement from the results of PPA-modified mixtures,
which correlates with the data obtained from the SFE analysis of the corresponding asphalt
binders.
0.00
1.00
2.00
3.00
4.00
5.00
Str
ength
(M
Pa)
Binder Type
S1B1 S1B3 S1B4 S1B7 S1B5
TRC 1501 Final Report
Page | 93
Figure 5.41: Summary of Tensile Strength Ratio.
5.4 Field Performance Data Collection
RAP samples containing PG 76-22 binder were collected from I-30 and I-40 to check the
presence of PPA along with any possible detrimental effects on the mixture performance. These
RAP samples were collected as part of a related project (ArDOT sponsored TRC 1404). In
addition, field performance data of these pavement sections were collected from the ArDOT
databases. The ArDOT-supplied MMHIS software was used to analyze the necessary data
retrieved from the database. Two types of pavement performance data (IRI and rutting) were
analyzed.
5.4.1 Acid Detection Test
At first, the recovered binders from RAPs of ten pavement sections were tested to detect the
presence of PPA. The binder was separated from aggregates of the RAP by following the
centrifuge method, in which nPB was used as the solvent. The binder from the solution
(binder+nPB) was then recovered by using a rotavapor. A total of 42 recovered binder samples
were tested for PPA detection. The details of these binder samples are shown in Table 5.10.
Among the 42 samples tested for acid detection test, only five of them were tested positive for
PPA. All five of those PPA-modified binders were recovered from one particular section. The
ArDOT job number for this section is B60115 (Section No. 10), which is located on I-30 near
Arkadelphia, AR. The cylindrical core samples were collected from the I-30 Westbound near
Mile Markers 242.8, 246.0, and 248.0. For these locations, the IRI and rutting data were also
0
5
10
15
20
25
30
S1B1 S1B3 S1B4 S1B7 S1B5
Ten
sile
Str
ength
, p
si
Binder Type
Wet Dry
TRC 1501 Final Report
Page | 94
analyzed by using the MMHIS software.
Table 5.10: Acid Detection Test Sample Details for Field Performance Data Collection
Section No. Section
Performance
Job number Number of
Samples Tested
Number of PPA
Detected Samples
1 Good B10102 9 0
2 Medium B10103 8 0
3 Poor B40102 7 0
4 Medium BX0103 5 0
5 Good BX0102 1 0
6 Poor BB0803 4 0
7 Poor BB0805 1 0
8 Poor BB0105 1 0
9 Good B70102 1 0
10 Good B60115 5 5
5.4.2 MMHIS Results
A higher roughness index indicates a severely deteriorated pavement, as roughness index is the
measure of the vertical stress received by the pavement surface. Moreover, a higher roughness
index indicates a pavement with more surface deformation than a pavement with a lower
roughness index. So, it is imperative that a pavement with good performance to have a lower
roughness index. Figure 5.42 shows the IRI data for Section 10, which is a “good” performing
section. The average IRI values indicate that the roughness of this section was well below the
threshold values for Arkansas.
The rutting data for this section is showed in Figure 5.43. Rutting is the permanent
surface depression, which is formed along the wheel paths. Similar to the IRI values, rutting
values for this section also met the ArDOT cutoff value for a good performing section. The
threshold value of rutting in Arkansas for a fair pavement is 0.35 in/mile. As Figure 5.43 shows
the rutting values stayed within this value for most of the times.
TRC 1501 Final Report
Page | 95
Figure 5.42: Average IRI Values for Mixture Containing PPA-modified Binder.
Figure 5.43: Average Rutting Data for Mixture Containing PPA.
5.4.3 Summary of Field Performance Data
The field performance of PPA-modified asphalt mixtures showed no causes of concerns, as the
section from which the PPA-modified binder was found was a good pavement section,
categorized by the ArDOT engineers.
0
20
40
60
80
100
120
140
160
201
3-1
0
20
13
-2
20
12
-3
2011-4
20
10
-3
20
09
-4
20
08
-6
2007
-11
2006C
20
05
-4
2005
-11
20
04
-8
20
04
-1
20
03
-4
20
03
-1
20
02
-6
20
02
-3
2001
IRI
(in
/mil
e)
Year and Month
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
2013
-10
2013-2
2012-3
2011-4
2010-3
2009-4
2008-6
2007-1
1
2006C
2005-4
2005-1
1
2004-8
2004-1
2003-4
2003-1
2002-6
2002-3
2001
Ru
ttin
g (
in/m
ile)
Year and Month
TRC 1501 Final Report
Page | 96
6. Conclusions and Recommendations
6.1 Conclusions
Penetration test results showed that PPA reduces the penetration value of asphalt binders,
which indicates that PPA hardens the binder. However, SBS-modified binders showed
even lower penetration values than the PPA-modified binders. Overall, the binders from
Source 2 (S2) were considerably harder than the binders from Source 1 (S1).
The rotational viscosity (RV) test results showed an increasing trend in viscosity of
asphalt binders after PPA modification. However, the SBS-modified binders showed
higher viscosities than the PPA-modified binders. The use of LAA decreased the
viscosity of PPA-modified PG 70-22 binder.
DSR test results showed an increase in G*/sinδ values in PPA-modified binders from
both sources under both unaged and RTFO aging conditions. However, in the case of
Source 1, all LAAs reduced the rutting factor of the PPA-modified binders. In the case of
Source 2, only Kao gripper X2 was able to maintain the rutting factor of PPA-modified
PG 70-22 binder.
The fatigue factor (G*sinδ) values from the DSR tests of PPA-modified binders were
less than the corresponding SBS-modified binders, which signifies the possibility of
better fatigue resistance of PPA-modified binder than the SBS-modified binder.
BBR test results showed that PPA modified binders showed lower m-values than
corresponding SBS-modified binders. Moreover, the PPA-modified binders showed
higher stiffness (S) values at a lower temperature than SBS-modified binder of the same
grade. This result could mean that PPA-modified binders could be susceptible to low
temperature cracking. Among the sixteen tested binder samples, 0.5% PPA modified
binder from Source 2 showed the lowest PG grade.
The SFE analysis showed PPA-modified asphalts had different cohesive energy for
binders from two different crude sources. And, the compatibility analysis of asphalt
binders with four different aggregates showed that PPA-modified asphalt binders from
Source 1 had lower adhesion energy in the dry condition and higher adhesion energy in
the wet condition. The compatibility analysis also showed higher CR values for PPA-
modified binders with the optimum dosage rates.
TRC 1501 Final Report
Page | 97
FTIR analysis of PPA-modified binders with or without LAA showed functional groups
such as alkyl halides, alkanes, hydroxyle and carboxylic acids. Throughout the spectrum,
various peaks were found which showed the presence of aromatics, which seem to be the
controlling functional group for the PPA-modified binders. Moreover, the spectrum of
the LAA binders did not show any changes functional groups, but they changed the
absorption values of the peaks.
Mixture test results revealed superior dynamic modulus, creep compliance and indirect
tensile strength, fracture energy, and tensile strength ratio of PPA-modified binders
compared to the SBS-modified binder of the same PG grade.
It is recommended that the dosage rate of PPA to be set upon the identification of the
crude source of the asphalt binder. If the source is unknown, a dosage rate between 0.5%
and 0.75% to be used.
6.2 Recommendations
In this study, only foamed based WMA technology was tested with PPA-modified
binders. However, the effects of different WMA additives needs to be tested on the
performance of PPA-modified asphalt binders to ensure the performance of PPA-
modified binders with WMA.
Several LAAs (mostly ArDOT certified ones) were tested for determining the effect with
PPA-modified binders. However, as the ArDOT allows many refineries outside Arkansas
to supply asphalt binders that might use different LAAs, which could have negative
results with PPA-modified binders. So further investigation of other LAAs in PPA-
modified asphalts would be helpful.
No RAP or RAS was used to modify asphalt binders in this study. So, the performance of
PPA-modified binders with RAP was unknown. The effects of PPA on RAP and RAS
need to be tested.
Multiple Stress Creep Recover (MSCR) testing is considered as the best approach for
evaluation of the rutting potential of the binder in place of the Elastic Recovery (ER).
Thus, it is recommended to conduct a thorough MSCR study to establish necessary
guidelines.
TRC 1501 Final Report
Page | 98
Since the ER (a PG Plus test) or MSCR test method is not capable of characterizing PPA-
modified binders, a new mechanistic or chemical based test method is recommended to
be explored to substitute them.
6.3 Implementation
As mentioned in Section 6.2, additional laboratory tests and field performance of PPA-modified
binders are recommended toward implementing the findings of this study. The following
implementation strategies can be explored while ArDOT undertakes further studies and/or
decisions on PPA-modified binders:
PPA can be considered as a potential modifier of asphalt binders.
It is suggested that ArDOT begins building a database about the combination of different
modifiers and crude sources of asphalt binders so that necessary directives can be provided to
the contractor in selecting compatible LAA and/or WMA additives. It would be the
contractor’s responsibility to obtain the information of the modifier used in the binder and
report it to the ArDOT to ensure compliance.
AASHTO TP 78 can be used to detect the presence of PPA in asphalt binder. Moreover,
Acidity measurement test should be used to determine the extent of acidity in asphalt binders.
The details of this test method would be available in the final report.
TRC 1501 Final Report
Page | 99
7 Acknowledgements
This document presents the preliminary recommendations for ArDOT of the research conducted
under project TRC 1501, Performance of Asphalt Modified with Polyphosphoric Acid. The
author would like to show their gratitude to ArDOT for providing financial assistance during the
course of this research project. The authors are grateful for the assistance of those who took part
in the research project, including ArDOT employees from the Systems Information and Research
Section and the State Materials Laboratory. The authors also acknowledge the talents and efforts
of graduate research assistants Istiaque Mahmud, Md. Shahriar Alam, Elvis Castillo, and Airam
Morales and undergraduate research assistants Shivakumar Ravishankar, Robert Darrington,
Leslie Parker, Anabella Monterroso, and Slater Smith. Industry partners, Ergon Inc. and Paragon
Technical Services are also greatly appreciated for their roles in the successful completion of this
project.
The statements contained in this report are the sole responsibility of the authors, and do
not necessarily reflect the official views of the Arkansas Department of Transportation. The use
of specific product names in this paper does not constitute or imply an endorsement of those
products. This document does not constitute a standard, specification, or regulation.
TRC 1501 Final Report
Page | 100
8 References
Walker, D. (2014). The Benefits of Modified Asphalts. Asphalt, 29(1).
Bahia, H. U., Hanson, D. I., Zeng, M., Zhai, H., Khatri, M. A., & Anderson, R. M. (2001).
Characterization of modified asphalt binders in superpave mix design (No. Project 9-10
FY'96).
D'Angelo, J., Kluttz, R., Dongre, R. N., Stephens, K., & Zanzotto, L. (2007). Revision of the
Superpave high temperature binder specification: The multiple stress creep recovery test
(With Discussion). Journal of the Association of Asphalt Paving Technologists, 76.
McDaniel, R., Soleymani, H., & Shah, A. (2002). Use of reclaimed asphalt pavement (RAP)
under Superpave specifications: A regional pooled fund project.
Yildirim, Y. (2007). Polymer modified asphalt binders. Construction and Building Materials,
21(1), 66-72.
Pederson, N. J. (2007). Pavement Lessons Learned from the AASHO Road Test and
Performance of the Interstate Highway System. Transport Research Board.
Monismith, C. L., Hudson, W. R., Finn, F. N., Skok Jr, E. L., & Shook, J. F. (2007). What
Pavement Research Was Done Following the AASHO Road Test and What Else Could
Have Been Done But Was Not. Interstate Highway System, 31.
Hallin, J. P. (2004). Development of the 2002 guide for the design of new and rehabilitated
pavement structures: Phase II. Report for National Cooperative Highway Research
Program, Transportation Research Board, National Research Council.
Terrel, R.L., and J.A. Epps. (1993). Using Additives and Modifiers in Hot Mixed Asphalt (Part
A), National Asphalt Pavement Association, Report No. Q1P114A.
Lewandowski, L. H. (1994). Polymer modification of paving asphalt binders. Rubber Chemistry
and Technology, 67(3), 447-480.
Orange, G., Martin, J. V., Menapace, A., Hemsley, M., & Baumgardner, G. L. (2004). Rutting
and moisture resistance of asphalt mixtures containing polymer and polyphosphoric acid
modified bitumen. Road materials and pavement design, 5(3), 323-354.
PATENT. (1997). WO 98/44047 (Ergon Inc.).
Orange, G., Dupuis, D., Martin, J. V., Farcas, F., Such, C., & Marcant, B. (2004). Chemical
modification of bitumen through polyphosphoric acid: properties-micro-structure
relationship. In PROCEEDINGS OF THE 3RD EURASPHALT AND EUROBITUME
CONGRESS HELD VIENNA, MAY 2004 (Vol. 1).
Martin, J. V., Orange, G., & Marcant, B. (2004, June). Effect of Polyphosphoric Acid on Aging
Behavior of Bituminous Binder. In the 41st Petersen Asphalt Research Conference,
Laramie, WY.
Bennert, T., & Martin, J. V. (2012). Polyphosphoric Acid in Combination with Styrene-
Butadiene-Styrene Block Copolymer. Polyphosphoric Acid Modification of Asphalt
Binders, 70.
TRC 1501 Final Report
Page | 101
Tam, K. (2009). Polyphosphoric Acid–Modified Asphalt Cement. Polyphosphoric Acid
Modification of Asphalt Binders, 109.
McGennis, R. B. (2009). Implementation of Polyphosphoric Acid Modification of Asphalt
Binders and Related Experience. Polyphosphoric Acid Modification of Asphalt Binders,
136.
Ryan, J., & D'Angelo, J. A. (2012). Wisconsin Approach to Binder Modification. Transportation
Research E-Circular, (E-C160).
Baumgardner, G. L. (2009). Why and How of Polyphosphoric Acid Modification?
Polyphosphoric Acid Modification of Asphalt Binders, 14.
Masson, J. F. (2008). Brief review of the chemistry of polyphosphoric acid (PPA) and bitumen.
Energy & Fuels, 22(4), 2637-2640.
Pamplona, T.F., Faxina, A.L. (2015). Effect of Polyphosphoric Acid on Asphalt binders with
different chemical composition. 94th Annual Transportation Research Board Meeting
Compendium Papers, Washington, D.C.
Li, X., Clyne, T., Reinke, G., Johnson, E. N., Gibson, N, Kutay, M. E. (2011). Laboratory
Evaluation of Asphalt Binders and Mixtures Containing Polyphosphoric Acid. 90th
Annual Transportation Research Board Meeting Compendium Papers, Washington, D.C.
Kodrat, I., Sohn, D., Hesp, S. (2007). Comparison of Polyphosphoric Acid-Modified Asphalt
Binders with Straight and Polymer Modified Materials. J. Transportation Research
Board, No. 1998, pp. 47-55.
Huang, S., Turner, T., Miknis, F., Thomas, K. (2008). Long Term Aging Characteristics of
Polyphosphoric Acid-Modified Asphalts. J. Transportation Research Board, No. 2051,
pp. 1-7.
Jafari, M., Babazadeh, A., Aflaki, S., (2015). Effects of Stress Levels on the Creep and Recovery
Behavior of Modified Asphalt Binders with the Same Continuous Performance Grades.
94th Annual Transportation Research Board Meeting Compendium Papers, Washington,
D.C.
Huang, S., Miknis, F., Schuster, W., Salmans, S., Farrar, M., Boysen, R. (2011). Rheological
and chemical properties of hydrated lime and polyphosphoric acid- modified asphalts
with long term aging. Journal of Materials in Civil Engineering. 23(5).
Baumgardner, G. L., Masson, J. F., Hardee, J. R., Menapace, A. M., & Williams, A. G. (2005).
Polyphosphoric acid modified asphalt: proposed mechanisms. Journal of the Association
of Asphalt Paving Technologists, 74, 283-305.
Daranga, C., Clopotel, C., Mofolasayo, A., & Bahia, H. U. (2009). Storage Stability and Effect
of Mineral Surface on Polyphosphoric Acid-Modified Asphalt Binders. In Transportation
Research Board 88th Annual Meeting (No. 09-2973).
Shulga, O., Maldonado, R., Romagosa, H., (2012). Effect of Foaming on Performance of Asphalt
Modified with Polyphosphoric Acid. Retrieved from: http://www.icl-
pp.com/SiteCollectionDocuments/Industrial/PPA%20in%20Asphalt/Application%20of%
20PPA%20in%20WMA.pdf . Last accessed 10/11/16.
TRC 1501 Final Report
Page | 102
Maine Department of Transportation (2013). Field Test of a Polyphosphoric Acid (PPA)
Modified Asphalt Binder on Rt. 1 in Perry. Technical Report 13-03. Retrieved from:
http://ntl.bts.gov/lib/52000/52500/52557/report1303f.pdf. Last accessed 7/28/15.
Federal Highway Administration (FHWA). (2012). The Use and Performance of Asphalt Binder
Modified with Polyphosphoric Acid (PPA). FHWA-HIF-012-030.
D’Angelo., J. A. (2009). Effect of polyphosphoric acid on asphalt binder properties.
Transportation Research Circular E-C160, Polyphosphoric Acid Modification of Asphalt
Binders, PP. 27-39.
Al-Qadi, I. L., Abauwad, I. M., Dhasmana, H., & Coenen, A. R. (2014). Effects of Various
Asphalt Binder Additives/Modifiers on Moisture-Susceptible Asphaltic Mixtures. Illinois
Center for Transportation.
Yan, K., Zhang, H., & Xu, H. (2013). Effect of polyphosphoric acid on physical properties,
chemical composition and morphology of bitumen. Construction and Building Materials,
47, 92-98.
Baldino, N., Gabriele, D., Rossi, C. O., Seta, L., Lupi, F. R., & Caputo, P. (2012). Low
temperature rheology of polyphosphoric acid (PPA) added bitumen. Construction and
Building Materials, 36, 592-596.
Xiao, F., Amirkhanian, S., Wang, H., & Hao, P. (2014). Rheological property investigations for
polymer and polyphosphoric acid modified asphalt binders at high temperatures.
Construction and Building Materials, 64, 316-323.
Zhang, F., & Yu, J. (2009). The research for high-performance SBR compound modified asphalt.
Construction and Building Materials, 24(3), 410-418.
Ryan, J., & Braham, A. (2015). The characterisation of foamed asphalt cement using a rotational
viscometer. International Journal of Pavement Engineering, 1-9.
Scribd (2016). Table of Characteristic IR Absorptions. Online Article. Retrieved from the
internet. https://www.scribd.com/doc/129244226/Table-of-Characteristic-IR-
Absorptions. Last Access 10/23/2016.
Jia, X., Huang, B., Bowers, B. F., & Zhao, S. (2014). Infrared spectra and rheological properties
of asphalt cement containing waste engine oil residues. Construction and Building
Materials, 50, 683-691.
Retrieved from the Internet. Characteristic IR Absorption Frequencies of Organic Functional
Groups. http://www2.ups.edu/faculty/hanson/Spectroscopy/IR/IRfrequencies.html. Last
Access 10/23/2016.
Bagampadde, U., & Isacsson, U. (2006). Characterization of chemical reactivity of liquid
antistripping additives using potentiometric titration and FTIR spectroscopy. Energy &
fuels, 20(5), 2174-2180.
Ghebremeskel, G. N., & Shield, S. R. (2003). Characterization of binary/tertiary blends of SBR,
NBR and PVC by IR spectroscopy. Rubber world, 227(4).
TRC 1501 Final Report
Page | 103
Cheng, D., Little, D., Lytton, R., & Holste, J. (2001). Surface energy measurement of asphalt and
its application to predicting fatigue and healing in asphalt mixtures. Transportation
Research Record: Journal of the Transportation Research Board, (1810), 44-53.
Van Oss, C. J., Chaudhury, M. K., and Good, R. J. (1998). Interfacial Lifshitz-van der Waals
and Polar Interactions in Macroscopic Systems. Chemical Reviews, 88(6), pp. 941-972.
Bahramian, A. (2012). Evaluating surface energy components of asphalt binders using Wilhelmy
Plate and Sessile Drop Techniques. Degree Project in Highway and Railway Engineering,
Stockholm, Sweden, 2002.
Buddhala, A., Edgar, A. O., Zaman, M., Wasiuddin, N. M., & Hossain, Z. (2011). Effects of an
amine anti-stripping agent on moisture susceptibility of sasobit and aspha-min mixes by
surface free energy analysis. Journal of Testing and Evaluation, 40(1), 1-9.
Wasiuddin, N. M., & Zaman, M. M. (2010). Polymeric aggregate treatment using styrene-
butadiene rubber (SBR) for moisture-induced damage potential. International Journal of
Pavement Research and Technology, 3(1), 1-9.
Kittu, A. T., Bulut, R., & Harimkar, S. (2013). Comparison of Contact Angle Measurements on
Granite with Respect to Surface Roughness.
Kittu, A. T., Bulut, R., & Harimkar, S. (2014, May). Comparison of Surface Energy Values of
Limestone with Respect to Different 3D Surface Roughness Measurements. In Pavement
Materials, Structures, and Performance (pp. 39-47). ASCE.
Koc, M., & Bulut, R. (2013). Evaluation of a warm mix asphalt additive using direct contact
angle measurements. In Int. J. Pavements Conf. 2013.
Koc, M., & Bulut, R. (2013). Assessment of a sessile drop device and a new testing approach
measuring contact angles on aggregates and asphalt binders. Journal of Materials in Civil
Engineering, 26(3), 391-398.
Little, D. N., Bhasin, A., & Hefer, A. W. (2006). Using surface energy measurements to select
materials for asphalt pavement. Texas Transportation Institute, Draft Final Report,
Project 9-37, National Cooperative Highway Research Program, Washington D.C., 2006
Wasiuddin, N. M., Fogle, C. M., Zaman, M. M., & O'Rear, E. A. (2007). Effect of antistrip
additives on surface free energy characteristics of asphalt binders for moisture-induced
damage potential. Journal of testing and evaluation, 35(1), 36-44.
Arabani, M., Roshani, H., & Hamedi, G. H. (2011). Estimating moisture sensitivity of warm mix
asphalt modified with zycosoil as an antistrip agent using surface free energy method.
Journal of Materials in Civil Engineering, 24(7), 889-897.
Hefer, A. W., Bhasin, A., & Little, D. N. (2005). Bitumen surface energy characterization using
a contact angle approach. Journal of Materials in Civil Engineering, 18(6), 759-767.
Yang, S., Braham, A., Chowdhury, N., & Hossain, Z. (2015, July). Linking the Field and Lab
Performance of Interstate Pavements. In International Symposium on Systematic
Approaches to Environmental Sustainability in Transportation.
Chowdhury, N., Hossain, Z., Yang, S., & Braham, A. (2015, March). A Framework for
Premature Pavement Distress Evaluation. In IFCEE 2015.